[PATCH v4 0/8] Replay Protected Memory Block (RPMB) subsystem

[PATCH v4 0/8] Replay Protected Memory Block (RPMB) subsystem

[Tomas Winkler]

Few storage technology such is EMMC, UFS, and NVMe support RPMB
hardware partition with common protocol and frame layout.
The RPMB partition cannot be accessed via standard block layer, but
by a set of specific commands: WRITE, READ, GET_WRITE_COUNTER, and
PROGRAM_KEY.
Such a partition provides authenticated and replay protected access,
hence suitable as a secure storage.

A storage device registers its RPMB hardware (emmc) partition or
RPMB W-LUN (ufs) with the RPMB layer providing an implementation for
send_rpmb_req() handler.
There is as well simulation platform device. This is handy as an RPMB
key can be programmed only once at storage device lifetime.

The RPMB layer aims to provide in-kernel API for Trusted Execution
Environment (TEE) devices that are capable to securely compute block
frame signature. In case a TEE device wish to store a replay protected
data, it creates an RPMB frame with requested data and computes HMAC of
the frame, then it requests the storage device via RPMB layer to store
the data.
A TEE driver can claim rpmb interface, for example,
via  class_interface_register ().

A parallel user space API is provided via /dev/rpmbX character
device with a single IOCTL command similar to the one provided by
mmc/ioctl. h
This API may help applications such as 
https://android.googlesource.com/trusty/app/storage/
To be obliviose to the underlaying HW storage technology.

There is a sample tool under tools/rpmb/ directory that exercises
this interface.

Tomas Winkler (8):
  rpmb: add Replay Protected Memory Block (RPMB) subsystem
  char: rpmb: add sysfs-class ABI documentation
  char: rpmb: add device attributes
  char: rpmb: provide user space interface
  char: rpmb: add RPMB simulation device
  tools rpmb: add RPBM access tool
  mmc: block: register RPMB partition with the RPMB subsystem
  scsi: ufs: connect to RPMB subsystem

 Documentation/ABI/testing/sysfs-class-rpmb |  44 ++
 Documentation/ioctl/ioctl-number.txt       |   1 +
 MAINTAINERS                                |  10 +
 drivers/char/Kconfig                       |   2 +
 drivers/char/Makefile                      |   1 +
 drivers/char/rpmb/Kconfig                  |  25 +
 drivers/char/rpmb/Makefile                 |   6 +
 drivers/char/rpmb/cdev.c                   | 207 +++++++
 drivers/char/rpmb/core.c                   | 408 ++++++++++++++
 drivers/char/rpmb/rpmb-cdev.h              |  31 ++
 drivers/char/rpmb/rpmb_sim.c               | 589 ++++++++++++++++++++
 drivers/mmc/card/Kconfig                   |   1 +
 drivers/mmc/card/block.c                   | 292 ++++++++++
 drivers/scsi/ufs/Kconfig                   |   1 +
 drivers/scsi/ufs/ufshcd.c                  | 219 ++++++++
 drivers/scsi/ufs/ufshcd.h                  |   2 +
 include/linux/rpmb.h                       | 138 +++++
 include/uapi/linux/rpmb.h                  | 120 ++++
 tools/Makefile                             |  14 +-
 tools/rpmb/.gitignore                      |   2 +
 tools/rpmb/Makefile                        |  32 ++
 tools/rpmb/rpmb.c                          | 862 +++++++++++++++++++++++++++++
 22 files changed, 3002 insertions(+), 5 deletions(-)
 create mode 100644 Documentation/ABI/testing/sysfs-class-rpmb
 create mode 100644 drivers/char/rpmb/Kconfig
 create mode 100644 drivers/char/rpmb/Makefile
 create mode 100644 drivers/char/rpmb/cdev.c
 create mode 100644 drivers/char/rpmb/core.c
 create mode 100644 drivers/char/rpmb/rpmb-cdev.h
 create mode 100644 drivers/char/rpmb/rpmb_sim.c
 create mode 100644 include/linux/rpmb.h
 create mode 100644 include/uapi/linux/rpmb.h
 create mode 100644 tools/rpmb/.gitignore
 create mode 100644 tools/rpmb/Makefile
 create mode 100644 tools/rpmb/rpmb.c

-- 
2.5.5

[PATCH v4 1/8] rpmb: add Replay Protected Memory Block (RPMB) subsystem

[Tomas Winkler]

Few storage technologies such is EMMC, UFS, and NVMe support RPMB hardware
partition with common protocol and frame layout.
The RPMB partition cannot be accessed via standard block layer, but by a
set of specific commands: WRITE, READ, GET_WRITE_COUNTER, and PROGRAM_KEY.
Such a partition provides authenticated and replay protected access,
hence suitable as a secure storage.

The RPMB layer aims to provide in-kernel API for Trusted Execution
Environment (TEE) devices that are capable to securely compute block
frame signature. In case a TEE device wish to store a replay protected
data, it creates an RPMB frame with requested data and computes HMAC of
the frame, then it requests the storage device via RPMB layer to store
the data.
A TEE driver can claim the RPMB interface, for example,
via class_interface_register ().

A storage device registers its RPMB hardware (eMMC) partition or
RPMB W-LUN (UFS) with the RPMB layer providing an implementation
for send_rpmb_req() handler.

V2: added short workflow description in the commit message
V3: commit message fix
V4: resend

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
---
 MAINTAINERS                |   7 +
 drivers/char/Kconfig       |   2 +
 drivers/char/Makefile      |   1 +
 drivers/char/rpmb/Kconfig  |   8 ++
 drivers/char/rpmb/Makefile |   4 +
 drivers/char/rpmb/core.c   | 337 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/rpmb.h       | 200 +++++++++++++++++++++++++++
 7 files changed, 559 insertions(+)
 create mode 100644 drivers/char/rpmb/Kconfig
 create mode 100644 drivers/char/rpmb/Makefile
 create mode 100644 drivers/char/rpmb/core.c
 create mode 100644 include/linux/rpmb.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 7304d2e37a98..e045c3a43920 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9691,6 +9691,13 @@ F:	include/net/rose.h
 F:	include/uapi/linux/rose.h
 F:	net/rose/
 
+RPMB SUBSYSTEM
+M:	Tomas Winkler <tomas.winkler@intel.com>
+L:	linux-kernel@vger.kernel.org
+S:	Supported
+F:	drivers/char/rpmb/*
+F:	include/linux/rpmb.h
+
 RTL2830 MEDIA DRIVER
 M:	Antti Palosaari <crope@iki.fi>
 L:	linux-media@vger.kernel.org
diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 601f64fcc890..2dc9733a6427 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -602,5 +602,7 @@ config TILE_SROM
 
 source "drivers/char/xillybus/Kconfig"
 
+source "drivers/char/rpmb/Kconfig"
+
 endmenu
 
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index d8a7579300d2..1a5e43c1a9a6 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -60,3 +60,4 @@ js-rtc-y = rtc.o
 
 obj-$(CONFIG_TILE_SROM)		+= tile-srom.o
 obj-$(CONFIG_XILLYBUS)		+= xillybus/
+obj-$(CONFIG_RPMB)		+= rpmb/
diff --git a/drivers/char/rpmb/Kconfig b/drivers/char/rpmb/Kconfig
new file mode 100644
index 000000000000..c5e6e909efce
--- /dev/null
+++ b/drivers/char/rpmb/Kconfig
@@ -0,0 +1,8 @@
+config RPMB
+	tristate "RPMB partition interface"
+	help
+	  Unified RPMB partition interface for eMMC and UFS.
+	  Provides interface for in kernel security controllers to
+	  access RPMB partition.
+
+	  If unsure, select N.
diff --git a/drivers/char/rpmb/Makefile b/drivers/char/rpmb/Makefile
new file mode 100644
index 000000000000..812b3ed264c0
--- /dev/null
+++ b/drivers/char/rpmb/Makefile
@@ -0,0 +1,4 @@
+obj-$(CONFIG_RPMB) += rpmb.o
+rpmb-objs += core.o
+
+ccflags-y += -D__CHECK_ENDIAN__
diff --git a/drivers/char/rpmb/core.c b/drivers/char/rpmb/core.c
new file mode 100644
index 000000000000..c9ea30aca6be
--- /dev/null
+++ b/drivers/char/rpmb/core.c
@@ -0,0 +1,337 @@
+/*
+ * Copyright (C) 2015-2016 Intel Corp. All rights reserved
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/mutex.h>
+#include <linux/list.h>
+#include <linux/device.h>
+#include <linux/slab.h>
+
+#include <linux/rpmb.h>
+
+static DEFINE_IDA(rpmb_ida);
+
+/**
+ * rpmb_dev_get - increase rpmb device ref counter
+ *
+ * @rdev: rpmb device
+ */
+struct rpmb_dev *rpmb_dev_get(struct rpmb_dev *rdev)
+{
+	return get_device(&rdev->dev) ? rdev : NULL;
+}
+EXPORT_SYMBOL_GPL(rpmb_dev_get);
+
+/**
+ * rpmb_dev_put - decrease rpmb device ref counter
+ *
+ * @rdev: rpmb device
+ */
+void rpmb_dev_put(struct rpmb_dev *rdev)
+{
+	put_device(&rdev->dev);
+}
+EXPORT_SYMBOL_GPL(rpmb_dev_put);
+
+static int rpmb_request_verify(struct rpmb_dev *rdev, struct rpmb_data *rpmbd)
+{
+	u16 req_type, block_count;
+
+	if (rpmbd->in_frames == NULL  || rpmbd->out_frames == NULL ||
+	    rpmbd->in_frames_cnt == 0 || rpmbd->out_frames_cnt == 0)
+		return -EINVAL;
+
+	req_type = be16_to_cpu(rpmbd->in_frames[0].req_resp);
+	block_count = be16_to_cpu(rpmbd->in_frames[0].block_count);
+
+	if (rpmbd->req_type != req_type) {
+		dev_err(&rdev->dev, "rpmb req type doesn't match 0x%04X = 0x%04X\n",
+			req_type, rpmbd->req_type);
+		return -EINVAL;
+	}
+
+	switch (req_type) {
+	case RPMB_PROGRAM_KEY:
+		dev_dbg(&rdev->dev, "rpmb program key = 0x%1x blk = %d\n",
+			req_type, block_count);
+		break;
+	case RPMB_GET_WRITE_COUNTER:
+		dev_dbg(&rdev->dev, "rpmb get write counter = 0x%1x blk = %d\n",
+			req_type, block_count);
+
+		break;
+	case RPMB_WRITE_DATA:
+		dev_dbg(&rdev->dev, "rpmb write data = 0x%1x blk = %d\n",
+			req_type, block_count);
+
+		if (rdev->ops->reliable_wr_cnt &&
+		    block_count > rdev->ops->reliable_wr_cnt) {
+			dev_err(&rdev->dev, "rpmb write data: block count %u > reliable wr count %u\n",
+				block_count, rdev->ops->reliable_wr_cnt);
+			return -EINVAL;
+		}
+
+		if (block_count > rpmbd->in_frames_cnt) {
+			dev_err(&rdev->dev, "rpmb write data: block count %u > in frame count %u\n",
+				block_count, rpmbd->in_frames_cnt);
+			return -EINVAL;
+		}
+		break;
+	case RPMB_READ_DATA:
+		dev_dbg(&rdev->dev, "rpmb read data = 0x%1x blk = %d\n",
+			req_type, block_count);
+
+		if (block_count > rpmbd->out_frames_cnt) {
+			dev_err(&rdev->dev, "rpmb read data: block count %u > out frame count %u\n",
+				block_count, rpmbd->in_frames_cnt);
+			return -EINVAL;
+		}
+		break;
+	case RPMB_RESULT_READ:
+		/* Internal command not supported */
+		dev_err(&rdev->dev, "NOTSUPPORTED rpmb resut read = 0x%1x blk = %d\n",
+			req_type, block_count);
+		return -EOPNOTSUPP;
+
+	default:
+		dev_err(&rdev->dev, "Error rpmb invalid command = 0x%1x blk = %d\n",
+			req_type, block_count);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * rpmb_send_req - send rpmb request
+ *
+ * @rdev: rpmb device
+ * @data: rpmb request data
+ *
+ * Return: 0 on success
+ *         -EINVAL on wrong parameters
+ *         -EOPNOTSUPP if device doesn't support the requested operation
+ *         < 0 if the operation fails
+ */
+int rpmb_send_req(struct rpmb_dev *rdev, struct rpmb_data *data)
+{
+	int err;
+
+	if (!rdev || !data)
+		return -EINVAL;
+
+	err = rpmb_request_verify(rdev, data);
+	if (err)
+		return err;
+
+	mutex_lock(&rdev->lock);
+	if (rdev->ops && rdev->ops->send_rpmb_req)
+		err = rdev->ops->send_rpmb_req(rdev->dev.parent, data);
+	else
+		err = -EOPNOTSUPP;
+	mutex_unlock(&rdev->lock);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rpmb_send_req);
+
+static void rpmb_dev_release(struct device *dev)
+{
+	struct rpmb_dev *rdev = to_rpmb_dev(dev);
+
+	ida_simple_remove(&rpmb_ida, rdev->id);
+	kfree(rdev);
+}
+
+struct class rpmb_class = {
+	.name = "rpmb",
+	.owner = THIS_MODULE,
+	.dev_release = rpmb_dev_release,
+};
+EXPORT_SYMBOL(rpmb_class);
+
+/**
+ * rpmb_dev_find_device - return first matching rpmb device
+ *
+ * @data: data for the match function
+ * @match: the matching function
+ *
+ * Return: matching rpmb device or NULL on failure
+ */
+struct rpmb_dev *rpmb_dev_find_device(void *data,
+			int (*match)(struct device *dev, const void *data))
+{
+	struct device *dev;
+
+	dev = class_find_device(&rpmb_class, NULL, data, match);
+
+	return dev ? to_rpmb_dev(dev) : NULL;
+}
+EXPORT_SYMBOL_GPL(rpmb_dev_find_device);
+
+static int match_by_type(struct device *dev, const void *data)
+{
+	struct rpmb_dev *rdev = to_rpmb_dev(dev);
+	enum rpmb_type *type = (enum rpmb_type *)data;
+
+	return (*type == RPMB_TYPE_ANY || rdev->ops->type == *type);
+}
+
+/**
+ * rpmb_dev_get_by_type - return first registered rpmb device
+ *      with matching type.
+ *      If run with RPMB_TYPE_ANY the first an probably only
+ *      device is returned
+ *
+ * @type: rpbm underlying device type
+ *
+ * Return: matching rpmb device or NULL/ERR_PTR on failure
+ */
+struct rpmb_dev *rpmb_dev_get_by_type(enum rpmb_type type)
+{
+	if (type > RPMB_TYPE_MAX)
+		return ERR_PTR(-EINVAL);
+
+	return rpmb_dev_find_device(&type, match_by_type);
+}
+EXPORT_SYMBOL_GPL(rpmb_dev_get_by_type);
+
+static int match_by_parent(struct device *dev, const void *data)
+{
+	const struct device *parent = data;
+
+	return (parent && dev->parent == parent);
+}
+
+/**
+ * rpmb_dev_find_by_device - retrieve rpmb device from the parent device
+ *
+ * @parent: parent device of the rpmb device
+ *
+ * Return: NULL if there is no rpmb device associated with the parent device
+ */
+struct rpmb_dev *rpmb_dev_find_by_device(struct device *parent)
+{
+	if (!parent)
+		return NULL;
+
+	return rpmb_dev_find_device(parent, match_by_parent);
+}
+EXPORT_SYMBOL_GPL(rpmb_dev_find_by_device);
+
+/**
+ * rpmb_dev_unregister - unregister RPMB partition from the RPMB subsystem
+ *
+ * @dev: parent device of the rpmb device
+ */
+int rpmb_dev_unregister(struct device *dev)
+{
+	struct rpmb_dev *rdev;
+
+	if (!dev)
+		return -EINVAL;
+
+	rdev = rpmb_dev_find_by_device(dev);
+	if (!rdev) {
+		dev_warn(dev, "no disk found %s\n", dev_name(dev->parent));
+		return -ENODEV;
+	}
+
+	rpmb_dev_put(rdev);
+
+	mutex_lock(&rdev->lock);
+	device_del(&rdev->dev);
+	mutex_unlock(&rdev->lock);
+
+	rpmb_dev_put(rdev);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(rpmb_dev_unregister);
+
+
+/**
+ * rpmb_dev_register - register RPMB partition with the RPMB subsystem
+ *
+ * @dev: storage device of the rpmb device
+ * @ops: device specific operations
+ */
+struct rpmb_dev *rpmb_dev_register(struct device *dev,
+				   const struct rpmb_ops *ops)
+{
+	struct rpmb_dev *rdev;
+	int id;
+	int ret;
+
+	if (!dev || !ops)
+		return ERR_PTR(-EINVAL);
+
+	if (!ops->send_rpmb_req)
+		return ERR_PTR(-EINVAL);
+
+	if (ops->type == RPMB_TYPE_ANY || ops->type > RPMB_TYPE_MAX)
+		return ERR_PTR(-EINVAL);
+
+	rdev = kzalloc(sizeof(struct rpmb_dev), GFP_KERNEL);
+	if (!rdev)
+		return ERR_PTR(-ENOMEM);
+
+	id = ida_simple_get(&rpmb_ida, 0, 0, GFP_KERNEL);
+	if (id < 0) {
+		ret = id;
+		goto exit;
+	}
+
+	mutex_init(&rdev->lock);
+	rdev->ops = ops;
+	rdev->id = id;
+
+	dev_set_name(&rdev->dev, "rpmb%d", id);
+	rdev->dev.class = &rpmb_class;
+	rdev->dev.parent = dev;
+	ret = device_register(&rdev->dev);
+	if (ret)
+		goto exit;
+
+	dev_dbg(&rdev->dev, "registered disk\n");
+
+	return rdev;
+
+exit:
+	if (id >= 0)
+		ida_simple_remove(&rpmb_ida, id);
+	kfree(rdev);
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL_GPL(rpmb_dev_register);
+
+static int __init rpmb_init(void)
+{
+	ida_init(&rpmb_ida);
+	class_register(&rpmb_class);
+	return 0;
+}
+
+static void __exit rpmb_exit(void)
+{
+	class_unregister(&rpmb_class);
+	ida_destroy(&rpmb_ida);
+}
+
+subsys_initcall(rpmb_init);
+module_exit(rpmb_exit);
+
+MODULE_AUTHOR("Intel Corporation");
+MODULE_DESCRIPTION("RPMB class");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/rpmb.h b/include/linux/rpmb.h
new file mode 100644
index 000000000000..2c1e259f1a04
--- /dev/null
+++ b/include/linux/rpmb.h
@@ -0,0 +1,200 @@
+/*
+ * Copyright (C) 2015-2016 Intel Corp. All rights reserved
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#ifndef __RPMB_H__
+#define __RPMB_H__
+
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/kref.h>
+
+/**
+ * struct rpmb_frame - rpmb frame as defined by specs
+ *
+ * @stuff        : stuff bytes
+ * @key_mac      : The authentication key or the message authentication
+ *                 code (MAC) depending on the request/response type.
+ *                 The MAC will be delivered in the last (or the only)
+ *                 block of data.
+ * @data         : Data to be written or read by signed access.
+ * @nonce        : Random number generated by the host for the requests
+ *                 and copied to the response by the RPMB engine.
+ * @write_counter: Counter value for the total amount of the successful
+ *                 authenticated data write requests made by the host.
+ * @addr         : Address of the data to be programmed to or read
+ *                 from the RPMB. Address is the serial number of
+ *                 the accessed block (half sector 256B).
+ * @block_count  : Number of blocks (half sectors, 256B) requested to be
+ *                 read/programmed.
+ * @result       : Includes information about the status of the write counter
+ *                 (valid, expired) and result of the access made to the RPMB.
+ * @req_resp     : Defines the type of request and response to/from the memory.
+ */
+struct rpmb_frame {
+	u8     stuff[196];
+	u8     key_mac[32];
+	u8     data[256];
+	u8     nonce[16];
+	__be32 write_counter;
+	__be16 addr;
+	__be16 block_count;
+	__be16 result;
+	__be16 req_resp;
+} __packed;
+
+#define RPMB_PROGRAM_KEY       0x1    /* Program RPMB Authentication Key */
+#define RPMB_GET_WRITE_COUNTER 0x2    /* Read RPMB write counter */
+#define RPMB_WRITE_DATA        0x3    /* Write data to RPMB partition */
+#define RPMB_READ_DATA         0x4    /* Read data from RPMB partition */
+#define RPMB_RESULT_READ       0x5    /* Read result request  (Internal) */
+
+#define RPMB_REQ2RESP(_OP) ((_OP) << 8)
+#define RPMB_RESP2REQ(_OP) ((_OP) >> 8)
+
+/**
+ * enum rpmb_op_result - rpmb operation results
+ *
+ * @RPMB_ERR_OK      : operation successful
+ * @RPMB_ERR_GENERAL : general failure
+ * @RPMB_ERR_AUTH    : mac doesn't match or ac calculation failure
+ * @RPMB_ERR_COUNTER : counter doesn't match or counter increment failure
+ * @RPMB_ERR_ADDRESS : address out of range or wrong address alignment
+ * @RPMB_ERR_WRITE   : data, counter, or result write failure
+ * @RPMB_ERR_READ    : data, counter, or result read failure
+ * @RPMB_ERR_NO_KEY  : authentication key not yet programmed
+ *
+ * @RPMB_ERR_COUNTER_EXPIRED:  counter expired
+ */
+enum rpmb_op_result {
+	RPMB_ERR_OK      = 0x0000,
+	RPMB_ERR_GENERAL = 0x0001,
+	RPMB_ERR_AUTH    = 0x0002,
+	RPMB_ERR_COUNTER = 0x0003,
+	RPMB_ERR_ADDRESS = 0x0004,
+	RPMB_ERR_WRITE   = 0x0005,
+	RPMB_ERR_READ    = 0x0006,
+	RPMB_ERR_NO_KEY  = 0x0007,
+
+	RPMB_ERR_COUNTER_EXPIRED = 0x0080
+};
+
+/**
+ * enum rpmb_type - type of underlaying storage technology
+ *
+ * @RPMB_TYPE_ANY   : any type used for search only
+ * @RPMB_TYPE_EMMC  : emmc (JESD84-B50.1)
+ * @RPMB_TYPE_UFS   : UFS (JESD220)
+ * @RPMB_TYPE_MAX   : upper sentinel
+ */
+enum rpmb_type {
+	RPMB_TYPE_ANY = 0,
+	RPMB_TYPE_EMMC,
+	RPMB_TYPE_UFS,
+	RPMB_TYPE_MAX = RPMB_TYPE_UFS
+};
+
+extern struct class rpmb_class;
+
+/**
+ * struct rpmb_data - rpmb data be transmitted in RPMB request
+ *
+ * @in_frames     :  list of input frames
+ * @out_frames    :  list of result frames
+ * @in_frames_cnt :  count of the input frames
+ * @out_frames_cnt:  count of the output frames
+ * @req_type      :  request type (program key, read, write, write counter)
+ */
+struct rpmb_data {
+	struct rpmb_frame *in_frames;
+	struct rpmb_frame *out_frames;
+	u32 in_frames_cnt;
+	u32 out_frames_cnt;
+	u16 req_type;
+};
+
+/**
+ * struct rpmb_ops - RPMB ops to be implemented by underlaying block device
+ *
+ * @send_rpmb_req  : send RPMB request to RPBM partition backed by the disk
+ * @type           : block device type
+ * @dev_id         : unique device identifier
+ * @dev_id_len     : unique device identifier length
+ * @reliable_wr_cnt: number of sectors that can be written in one access
+ */
+struct rpmb_ops {
+	int (*send_rpmb_req)(struct device *dev, struct rpmb_data *req);
+	enum rpmb_type type;
+	const u8 *dev_id;
+	size_t dev_id_len;
+	u16 reliable_wr_cnt;
+};
+
+/**
+ * struct rpmb_dev - device which can support RPMB partition
+ *
+ * @lock       : lock
+ * @dev        : device
+ * @id         : device id
+ * @ops        : operation exported by block layer
+ */
+struct rpmb_dev {
+	struct mutex lock;
+	struct device dev;
+	int    id;
+	const struct rpmb_ops *ops;
+};
+
+#define to_rpmb_dev(x) container_of((x), struct rpmb_dev, dev)
+
+#if IS_ENABLED(CONFIG_RPMB)
+struct rpmb_dev *rpmb_dev_get(struct rpmb_dev *rdev);
+void rpmb_dev_put(struct rpmb_dev *rdev);
+struct rpmb_dev *rpmb_dev_find_by_device(struct device *parent);
+struct rpmb_dev *rpmb_dev_get_by_type(enum rpmb_type type);
+struct rpmb_dev *rpmb_dev_register(struct device *dev,
+				   const struct rpmb_ops *ops);
+struct rpmb_dev *rpmb_dev_find_device(void *data,
+			int (*match)(struct device *dev, const void *data));
+int rpmb_dev_unregister(struct device *dev);
+int rpmb_send_req(struct rpmb_dev *rdev, struct rpmb_data *data);
+
+#else
+static inline struct rpmb_dev *rpmb_dev_get(struct rpmb_dev *rdev)
+{
+	return NULL;
+}
+static inline void rpmb_dev_put(struct rpmb_dev *rdev) { }
+static inline struct rpmb_dev *rpmb_dev_find_by_device(struct device *parent)
+{
+	return NULL;
+}
+static inline
+struct rpmb_dev *rpmb_dev_get_by_type(enum rpmb_type type)
+{
+	return NULL;
+}
+static inline struct rpmb_dev *
+rpmb_dev_register(struct device *dev, const struct rpmb_ops *ops)
+{
+	return NULL;
+}
+static inline int rpmb_dev_unregister(struct device *dev)
+{
+	return 0;
+}
+static inline int rpmb_send_req(struct rpmb_dev *rdev, struct rpmb_data *data)
+{
+	return 0;
+}
+#endif /* CONFIG_RPMB */
+
+#endif /* __RPMB_H__ */
-- 
2.5.5

[PATCH v4 2/8] char: rpmb: add sysfs-class ABI documentation

[Tomas Winkler]

V2: resend
V3: add more verbose description
V4: update kernel version to 4.8

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
 Documentation/ABI/testing/sysfs-class-rpmb | 20 ++++++++++++++++++++
 MAINTAINERS                                |  1 +
 2 files changed, 21 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-class-rpmb

diff --git a/Documentation/ABI/testing/sysfs-class-rpmb b/Documentation/ABI/testing/sysfs-class-rpmb
new file mode 100644
index 000000000000..44ee4d31dc0b
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-class-rpmb
@@ -0,0 +1,20 @@
+What:		/sys/class/rpmb/
+Date:		Jun 2016
+KernelVersion:  4.8
+Contact:	Tomas Winkler <tomas.winkler@intel.com>
+Description:
+		The rpmb/ class sub-directory belongs to RPMB device class.
+
+		Few storage technologies such is EMMC, UFS, and NVMe support
+		Replay Protected Memory Block (RPMB) hardware partition with
+		common protocol and frame layout.
+		Such a partition provides authenticated and replay protected access,
+		hence suitable as a secure storage.
+
+What:		/sys/class/rpmb/rpmbN/
+Date:		Jun 2016
+KernelVersion:	4.8
+Contact:	Tomas Winkler <tomas.winkler@intel.com>
+Description:
+		The /sys/class/rpmb/rpmbN directory is created for
+		each RPMB registered device
diff --git a/MAINTAINERS b/MAINTAINERS
index e045c3a43920..b4228dd58e06 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9697,6 +9697,7 @@ L:	linux-kernel@vger.kernel.org
 S:	Supported
 F:	drivers/char/rpmb/*
 F:	include/linux/rpmb.h
+F:	Documentation/ABI/testing/sysfs-class-rpmb
 
 RTL2830 MEDIA DRIVER
 M:	Antti Palosaari <crope@iki.fi>
-- 
2.5.5

[PATCH v4 3/8] char: rpmb: add device attributes

[Tomas Winkler]

Add attribute type that displays underlay storage type technology
EMMC, UFS, and attribute id, that displays underlay storage device id.
For EMMC this would be content of CID and for UFS serial number from
the device descriptor.

V2: resend
V3: set the kernel version to 4.7
V4: set the kernel version to 4.8

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
 Documentation/ABI/testing/sysfs-class-rpmb | 24 +++++++++++
 drivers/char/rpmb/core.c                   | 64 ++++++++++++++++++++++++++++++
 2 files changed, 88 insertions(+)

diff --git a/Documentation/ABI/testing/sysfs-class-rpmb b/Documentation/ABI/testing/sysfs-class-rpmb
index 44ee4d31dc0b..e9027f37ca5b 100644
--- a/Documentation/ABI/testing/sysfs-class-rpmb
+++ b/Documentation/ABI/testing/sysfs-class-rpmb
@@ -18,3 +18,27 @@ Contact:	Tomas Winkler <tomas.winkler@intel.com>
 Description:
 		The /sys/class/rpmb/rpmbN directory is created for
 		each RPMB registered device
+
+What:		/sys/class/rpmb/rpmbN/id
+Date:		Maj 2016
+KernelVersion:	4.7
+Contact:	Tomas Winkler <tomas.winkler@intel.com>
+Description:
+		The /sys/class/rpmb/rpmbN/id file contains device id
+		in a binary form
+
+What:		/sys/class/rpmb/rpmbN/type
+Date:		Jun 2016
+KernelVersion:	4.8
+Contact:	Tomas Winkler <tomas.winkler@intel.com>
+Description:
+		The /sys/class/rpmb/rpmbN/type file contains device
+		underlay storage type technology: EMMC, UFS
+
+What:		/sys/class/rpmb/rpmbN/reliable_wr_cnt
+Date:		Jun 2016
+KernelVersion:	4.8
+Contact:	Tomas Winkler <tomas.winkler@intel.com>
+Description:
+		The /sys/class/rpmb/rpmbN/reliable_wr_cnt file contains
+		number of blocks that can be written in a single request
diff --git a/drivers/char/rpmb/core.c b/drivers/char/rpmb/core.c
index c9ea30aca6be..fb401331d345 100644
--- a/drivers/char/rpmb/core.c
+++ b/drivers/char/rpmb/core.c
@@ -230,6 +230,68 @@ struct rpmb_dev *rpmb_dev_find_by_device(struct device *parent)
 }
 EXPORT_SYMBOL_GPL(rpmb_dev_find_by_device);
 
+static ssize_t type_show(struct device *dev,
+			 struct device_attribute *attr, char *buf)
+{
+	struct rpmb_dev *rdev = to_rpmb_dev(dev);
+	ssize_t ret;
+
+	switch (rdev->ops->type) {
+	case RPMB_TYPE_EMMC:
+		ret = scnprintf(buf, PAGE_SIZE, "EMMC\n");
+		break;
+	case RPMB_TYPE_UFS:
+		ret = scnprintf(buf, PAGE_SIZE, "UFS\n");
+		break;
+	default:
+		ret = scnprintf(buf, PAGE_SIZE, "UNKNOWN\n");
+		break;
+	}
+
+	return ret;
+}
+static DEVICE_ATTR_RO(type);
+
+static ssize_t id_show(struct device *dev,
+		       struct device_attribute *attr, char *buf)
+{
+	struct rpmb_dev *rdev = to_rpmb_dev(dev);
+	size_t sz = min_t(size_t, rdev->ops->dev_id_len, PAGE_SIZE);
+
+	if (!rdev->ops->dev_id)
+		return 0;
+
+	memcpy(buf, rdev->ops->dev_id, sz);
+	return sz;
+}
+static DEVICE_ATTR_RO(id);
+
+static ssize_t reliable_wr_cnt_show(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	struct rpmb_dev *rdev = to_rpmb_dev(dev);
+
+	return scnprintf(buf, PAGE_SIZE, "%u\n", rdev->ops->reliable_wr_cnt);
+}
+static DEVICE_ATTR_RO(reliable_wr_cnt);
+
+
+static struct attribute *rpmb_attrs[] = {
+	&dev_attr_type.attr,
+	&dev_attr_id.attr,
+	&dev_attr_reliable_wr_cnt.attr,
+	NULL,
+};
+
+static struct attribute_group rpmb_attr_group = {
+	.attrs = rpmb_attrs,
+};
+
+static const struct attribute_group *rpmb_attr_groups[] = {
+	&rpmb_attr_group,
+	NULL
+};
+
 /**
  * rpmb_dev_unregister - unregister RPMB partition from the RPMB subsystem
  *
@@ -300,6 +362,8 @@ struct rpmb_dev *rpmb_dev_register(struct device *dev,
 	dev_set_name(&rdev->dev, "rpmb%d", id);
 	rdev->dev.class = &rpmb_class;
 	rdev->dev.parent = dev;
+	rdev->dev.groups = rpmb_attr_groups;
+
 	ret = device_register(&rdev->dev);
 	if (ret)
 		goto exit;
-- 
2.5.5

[PATCH v4 4/8] char: rpmb: provide user space interface

[Tomas Winkler]

The user space API is achieved via single synchronous IOCTL.
The request is submitted in in_frames_ptr pointer and received
in out_frames_ptr.

V2: use memdup_user
V3: commit message fix
V4: resend

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
 Documentation/ioctl/ioctl-number.txt |   1 +
 MAINTAINERS                          |   1 +
 drivers/char/rpmb/Kconfig            |   7 ++
 drivers/char/rpmb/Makefile           |   1 +
 drivers/char/rpmb/cdev.c             | 207 +++++++++++++++++++++++++++++++++++
 drivers/char/rpmb/core.c             |   9 +-
 drivers/char/rpmb/rpmb-cdev.h        |  31 ++++++
 include/linux/rpmb.h                 |  78 ++-----------
 include/uapi/linux/rpmb.h            | 120 ++++++++++++++++++++
 9 files changed, 384 insertions(+), 71 deletions(-)
 create mode 100644 drivers/char/rpmb/cdev.c
 create mode 100644 drivers/char/rpmb/rpmb-cdev.h
 create mode 100644 include/uapi/linux/rpmb.h

diff --git a/Documentation/ioctl/ioctl-number.txt b/Documentation/ioctl/ioctl-number.txt
index 9369d3b0f09a..51a514befa48 100644
--- a/Documentation/ioctl/ioctl-number.txt
+++ b/Documentation/ioctl/ioctl-number.txt
@@ -320,6 +320,7 @@ Code  Seq#(hex)	Include File		Comments
 0xB1	00-1F	PPPoX			<mailto:mostrows@styx.uwaterloo.ca>
 0xB3	00	linux/mmc/ioctl.h
 0xB4	00-0F	linux/gpio.h		<mailto:linux-gpio@vger.kernel.org>
+0xB5	00	linux/uapi/linux/rpmb.h <mailto:linux-mei@linux.intel.com>
 0xC0	00-0F	linux/usb/iowarrior.h
 0xCA	00-0F	uapi/misc/cxl.h
 0xCA	80-8F	uapi/scsi/cxlflash_ioctl.h
diff --git a/MAINTAINERS b/MAINTAINERS
index b4228dd58e06..bf806f2d3e19 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9696,6 +9696,7 @@ M:	Tomas Winkler <tomas.winkler@intel.com>
 L:	linux-kernel@vger.kernel.org
 S:	Supported
 F:	drivers/char/rpmb/*
+F:	include/uapi/linux/rpmb.h
 F:	include/linux/rpmb.h
 F:	Documentation/ABI/testing/sysfs-class-rpmb
 
diff --git a/drivers/char/rpmb/Kconfig b/drivers/char/rpmb/Kconfig
index c5e6e909efce..6794be9fcc5e 100644
--- a/drivers/char/rpmb/Kconfig
+++ b/drivers/char/rpmb/Kconfig
@@ -6,3 +6,10 @@ config RPMB
 	  access RPMB partition.
 
 	  If unsure, select N.
+
+config RPMB_INTF_DEV
+	bool "RPMB character device interface /dev/rpmbN"
+	depends on RPMB
+	help
+	  Say yes here if you want to access RPMB from user space
+	  via character device interface /dev/rpmb%d
diff --git a/drivers/char/rpmb/Makefile b/drivers/char/rpmb/Makefile
index 812b3ed264c0..b5dc087b1299 100644
--- a/drivers/char/rpmb/Makefile
+++ b/drivers/char/rpmb/Makefile
@@ -1,4 +1,5 @@
 obj-$(CONFIG_RPMB) += rpmb.o
 rpmb-objs += core.o
+rpmb-$(CONFIG_RPMB_INTF_DEV) += cdev.o
 
 ccflags-y += -D__CHECK_ENDIAN__
diff --git a/drivers/char/rpmb/cdev.c b/drivers/char/rpmb/cdev.c
new file mode 100644
index 000000000000..67e94943d3e3
--- /dev/null
+++ b/drivers/char/rpmb/cdev.c
@@ -0,0 +1,207 @@
+/*
+ * Copyright (C) 2015-2016 Intel Corp. All rights reserved
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/uaccess.h>
+#include <linux/compat.h>
+#include <linux/slab.h>
+#include <linux/rpmb.h>
+
+#include "rpmb-cdev.h"
+
+static dev_t rpmb_devt;
+#define RPMB_MAX_DEVS  MINORMASK
+
+#define RPMB_DEV_OPEN    0  /** single open bit (position) */
+
+/**
+ * rpmb_open - the open function
+ *
+ * @inode: pointer to inode structure
+ * @file: pointer to file structure
+ *
+ * Return: 0 on success, <0 on error
+ */
+static int rpmb_open(struct inode *inode, struct file *file)
+{
+	struct rpmb_dev *rdev;
+
+	rdev = container_of(inode->i_cdev, struct rpmb_dev, cdev);
+	if (!rdev)
+		return -ENODEV;
+
+	/* the rpmb is single open! */
+	if (test_and_set_bit(RPMB_DEV_OPEN, &rdev->status))
+		return -EBUSY;
+
+	mutex_lock(&rdev->lock);
+
+	file->private_data = rdev;
+
+	mutex_unlock(&rdev->lock);
+
+	return nonseekable_open(inode, file);
+}
+
+static int rpmb_release(struct inode *inode, struct file *file)
+{
+	struct rpmb_dev *rdev = file->private_data;
+
+	clear_bit(RPMB_DEV_OPEN, &rdev->status);
+
+	return 0;
+}
+
+/*
+ * FIMXE: will be exported by the kernel in future version
+ * helper to convert user pointers passed inside __aligned_u64 fields
+ */
+static void __user *u64_to_ptr(__u64 val)
+{
+	return (void __user *) (unsigned long) val;
+}
+
+static int rpmb_ioctl_cmd(struct rpmb_dev *rdev,
+			  struct rpmb_ioc_cmd __user *ptr)
+{
+	struct rpmb_ioc_cmd cmd;
+	struct rpmb_data rpmbd;
+	struct rpmb_frame *in_frames = NULL;
+	struct rpmb_frame *out_frames = NULL;
+	size_t in_sz, out_sz;
+	int ret;
+
+	if (copy_from_user(&cmd, ptr, sizeof(cmd)))
+		return -EFAULT;
+
+	if (cmd.in_frames_count == 0 || cmd.out_frames_count == 0)
+		return -EINVAL;
+
+	in_sz = sizeof(struct rpmb_frame) * cmd.in_frames_count;
+	in_frames = memdup_user(u64_to_ptr(cmd.in_frames_ptr), in_sz);
+	if (IS_ERR(in_frames)) {
+		ret = PTR_ERR(in_frames);
+		in_frames = NULL;
+		goto out;
+	}
+
+	out_sz = sizeof(struct rpmb_frame) * cmd.out_frames_count;
+	out_frames = kmalloc(out_sz, GFP_KERNEL);
+	if (!out_frames) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	rpmbd.req_type = cmd.req;
+	rpmbd.in_frames = in_frames;
+	rpmbd.out_frames = out_frames;
+	rpmbd.in_frames_cnt = cmd.in_frames_count;
+	rpmbd.out_frames_cnt = cmd.out_frames_count;
+
+	ret  = rpmb_send_req(rdev, &rpmbd);
+	if (ret) {
+		dev_err(&rdev->dev, "Failed to process request = %d.\n", ret);
+		goto out;
+	}
+
+	if (copy_to_user(u64_to_ptr(cmd.out_frames_ptr), out_frames, out_sz)) {
+		ret = -EFAULT;
+		goto out;
+	}
+
+	if (copy_to_user(ptr, &cmd, sizeof(cmd))) {
+		ret = -EFAULT;
+		goto out;
+	}
+out:
+	kfree(in_frames);
+	kfree(out_frames);
+	return ret;
+}
+
+static long rpmb_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct rpmb_dev *rdev = file->private_data;
+	struct rpmb_ioc_cmd __user *req = (void __user *)arg;
+
+	if (cmd != RPMB_IOC_REQ) {
+		dev_err(&rdev->dev, "unsupported ioctl 0x%x.\n", cmd);
+		return -ENOIOCTLCMD;
+	}
+
+	return rpmb_ioctl_cmd(rdev, req);
+}
+
+#ifdef CONFIG_COMPAT
+static long compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct rpmb_dev *rdev = file->private_data;
+	struct rpmb_ioc_cmd __user *req = (void __user *)compat_ptr(arg);
+
+	if (cmd != RPMB_IOC_REQ) {
+		dev_err(&rdev->dev, "unsupported ioctl 0x%x.\n", cmd);
+		return -ENOIOCTLCMD;
+	}
+
+	return rpmb_ioctl_cmd(rdev, req);
+}
+#endif /* CONFIG_COMPAT */
+
+static const struct file_operations rpmb_fops = {
+	.open           = rpmb_open,
+	.release        = rpmb_release,
+	.unlocked_ioctl = rpmb_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl   = compat_ioctl,
+#endif
+	.owner          = THIS_MODULE,
+	.llseek         = noop_llseek,
+};
+
+void rpmb_cdev_prepare(struct rpmb_dev *rdev)
+{
+	rdev->dev.devt = MKDEV(MAJOR(rpmb_devt), rdev->id);
+	rdev->cdev.owner = THIS_MODULE;
+	cdev_init(&rdev->cdev, &rpmb_fops);
+}
+
+void rpmb_cdev_add(struct rpmb_dev *rdev)
+{
+	cdev_add(&rdev->cdev, rdev->dev.devt, 1);
+}
+
+void rpmb_cdev_del(struct rpmb_dev *rdev)
+{
+	if (rdev->dev.devt)
+		cdev_del(&rdev->cdev);
+}
+
+int __init rpmb_cdev_init(void)
+{
+	int ret;
+
+	ret = alloc_chrdev_region(&rpmb_devt, 0, RPMB_MAX_DEVS, "rpmb");
+	if (ret < 0)
+		pr_err("unable to allocate char dev region\n");
+
+	return ret;
+}
+
+void __exit rpmb_cdev_exit(void)
+{
+	if (rpmb_devt)
+		unregister_chrdev_region(rpmb_devt, RPMB_MAX_DEVS);
+}
diff --git a/drivers/char/rpmb/core.c b/drivers/char/rpmb/core.c
index fb401331d345..ce875d773ee4 100644
--- a/drivers/char/rpmb/core.c
+++ b/drivers/char/rpmb/core.c
@@ -20,6 +20,7 @@
 #include <linux/slab.h>
 
 #include <linux/rpmb.h>
+#include "rpmb-cdev.h"
 
 static DEFINE_IDA(rpmb_ida);
 
@@ -313,6 +314,7 @@ int rpmb_dev_unregister(struct device *dev)
 	rpmb_dev_put(rdev);
 
 	mutex_lock(&rdev->lock);
+	rpmb_cdev_del(rdev);
 	device_del(&rdev->dev);
 	mutex_unlock(&rdev->lock);
 
@@ -364,10 +366,14 @@ struct rpmb_dev *rpmb_dev_register(struct device *dev,
 	rdev->dev.parent = dev;
 	rdev->dev.groups = rpmb_attr_groups;
 
+	rpmb_cdev_prepare(rdev);
+
 	ret = device_register(&rdev->dev);
 	if (ret)
 		goto exit;
 
+	rpmb_cdev_add(rdev);
+
 	dev_dbg(&rdev->dev, "registered disk\n");
 
 	return rdev;
@@ -384,11 +390,12 @@ static int __init rpmb_init(void)
 {
 	ida_init(&rpmb_ida);
 	class_register(&rpmb_class);
-	return 0;
+	return rpmb_cdev_init();
 }
 
 static void __exit rpmb_exit(void)
 {
+	rpmb_cdev_exit();
 	class_unregister(&rpmb_class);
 	ida_destroy(&rpmb_ida);
 }
diff --git a/drivers/char/rpmb/rpmb-cdev.h b/drivers/char/rpmb/rpmb-cdev.h
new file mode 100644
index 000000000000..d20e0f16f1a1
--- /dev/null
+++ b/drivers/char/rpmb/rpmb-cdev.h
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2015-2016 Intel Corp. All rights reserved
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+
+
+#ifdef CONFIG_RPMB_INTF_DEV
+int __init rpmb_cdev_init(void);
+void __exit rpmb_cdev_exit(void);
+void rpmb_cdev_prepare(struct rpmb_dev *rdev);
+void rpmb_cdev_add(struct rpmb_dev *rdev);
+void rpmb_cdev_del(struct rpmb_dev *rdev);
+
+#else
+static inline int __init rpmb_cdev_init(void)
+{
+	return 0;
+}
+static inline void __exit rpmb_cdev_exit(void) {}
+static inline void rpmb_cdev_prepare(struct rpmb_dev *rdev) {}
+static inline void rpmb_cdev_add(struct rpmb_dev *rdev) {}
+static inline void rpmb_cdev_del(struct rpmb_dev *rdev) {}
+#endif /* CONFIG_RPMB_INTF_DEV */
diff --git a/include/linux/rpmb.h b/include/linux/rpmb.h
index 2c1e259f1a04..446c6639d275 100644
--- a/include/linux/rpmb.h
+++ b/include/linux/rpmb.h
@@ -15,79 +15,11 @@
 
 #include <linux/types.h>
 #include <linux/device.h>
+#include <linux/cdev.h>
+#include <uapi/linux/rpmb.h>
 #include <linux/kref.h>
 
 /**
- * struct rpmb_frame - rpmb frame as defined by specs
- *
- * @stuff        : stuff bytes
- * @key_mac      : The authentication key or the message authentication
- *                 code (MAC) depending on the request/response type.
- *                 The MAC will be delivered in the last (or the only)
- *                 block of data.
- * @data         : Data to be written or read by signed access.
- * @nonce        : Random number generated by the host for the requests
- *                 and copied to the response by the RPMB engine.
- * @write_counter: Counter value for the total amount of the successful
- *                 authenticated data write requests made by the host.
- * @addr         : Address of the data to be programmed to or read
- *                 from the RPMB. Address is the serial number of
- *                 the accessed block (half sector 256B).
- * @block_count  : Number of blocks (half sectors, 256B) requested to be
- *                 read/programmed.
- * @result       : Includes information about the status of the write counter
- *                 (valid, expired) and result of the access made to the RPMB.
- * @req_resp     : Defines the type of request and response to/from the memory.
- */
-struct rpmb_frame {
-	u8     stuff[196];
-	u8     key_mac[32];
-	u8     data[256];
-	u8     nonce[16];
-	__be32 write_counter;
-	__be16 addr;
-	__be16 block_count;
-	__be16 result;
-	__be16 req_resp;
-} __packed;
-
-#define RPMB_PROGRAM_KEY       0x1    /* Program RPMB Authentication Key */
-#define RPMB_GET_WRITE_COUNTER 0x2    /* Read RPMB write counter */
-#define RPMB_WRITE_DATA        0x3    /* Write data to RPMB partition */
-#define RPMB_READ_DATA         0x4    /* Read data from RPMB partition */
-#define RPMB_RESULT_READ       0x5    /* Read result request  (Internal) */
-
-#define RPMB_REQ2RESP(_OP) ((_OP) << 8)
-#define RPMB_RESP2REQ(_OP) ((_OP) >> 8)
-
-/**
- * enum rpmb_op_result - rpmb operation results
- *
- * @RPMB_ERR_OK      : operation successful
- * @RPMB_ERR_GENERAL : general failure
- * @RPMB_ERR_AUTH    : mac doesn't match or ac calculation failure
- * @RPMB_ERR_COUNTER : counter doesn't match or counter increment failure
- * @RPMB_ERR_ADDRESS : address out of range or wrong address alignment
- * @RPMB_ERR_WRITE   : data, counter, or result write failure
- * @RPMB_ERR_READ    : data, counter, or result read failure
- * @RPMB_ERR_NO_KEY  : authentication key not yet programmed
- *
- * @RPMB_ERR_COUNTER_EXPIRED:  counter expired
- */
-enum rpmb_op_result {
-	RPMB_ERR_OK      = 0x0000,
-	RPMB_ERR_GENERAL = 0x0001,
-	RPMB_ERR_AUTH    = 0x0002,
-	RPMB_ERR_COUNTER = 0x0003,
-	RPMB_ERR_ADDRESS = 0x0004,
-	RPMB_ERR_WRITE   = 0x0005,
-	RPMB_ERR_READ    = 0x0006,
-	RPMB_ERR_NO_KEY  = 0x0007,
-
-	RPMB_ERR_COUNTER_EXPIRED = 0x0080
-};
-
-/**
  * enum rpmb_type - type of underlaying storage technology
  *
  * @RPMB_TYPE_ANY   : any type used for search only
@@ -144,12 +76,18 @@ struct rpmb_ops {
  * @lock       : lock
  * @dev        : device
  * @id         : device id
+ * @cdev       : character dev
+ * @status     : device status
  * @ops        : operation exported by block layer
  */
 struct rpmb_dev {
 	struct mutex lock;
 	struct device dev;
 	int    id;
+#ifdef CONFIG_RPMB_INTF_DEV
+	struct cdev cdev;
+	unsigned long status;
+#endif /* CONFIG_RPMB_INTF_DEV */
 	const struct rpmb_ops *ops;
 };
 
diff --git a/include/uapi/linux/rpmb.h b/include/uapi/linux/rpmb.h
new file mode 100644
index 000000000000..55aa92e11280
--- /dev/null
+++ b/include/uapi/linux/rpmb.h
@@ -0,0 +1,120 @@
+/*
+ * Copyright (C) 2015-2016, Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation;
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef _UAPI_LINUX_RPMB_H_
+#define _UAPI_LINUX_RPMB_H_
+
+#include <linux/types.h>
+
+/**
+ * struct rpmb_frame - rpmb frame as defined by specs
+ *
+ * @stuff        : stuff bytes
+ * @key_mac      : The authentication key or the message authentication
+ *                 code (MAC) depending on the request/response type.
+ *                 The MAC will be delivered in the last (or the only)
+ *                 block of data.
+ * @data         : Data to be written or read by signed access.
+ * @nonce        : Random number generated by the host for the requests
+ *                 and copied to the response by the RPMB engine.
+ * @write_counter: Counter value for the total amount of the successful
+ *                 authenticated data write requests made by the host.
+ * @addr         : Address of the data to be programmed to or read
+ *                 from the RPMB. Address is the serial number of
+ *                 the accessed block (half sector 256B).
+ * @block_count  : Number of blocks (half sectors, 256B) requested to be
+ *                 read/programmed.
+ * @result       : Includes information about the status of the write counter
+ *                 (valid, expired) and result of the access made to the RPMB.
+ * @req_resp     : Defines the type of request and response to/from the memory.
+ */
+struct rpmb_frame {
+	__u8   stuff[196];
+	__u8   key_mac[32];
+	__u8   data[256];
+	__u8   nonce[16];
+	__be32 write_counter;
+	__be16 addr;
+	__be16 block_count;
+	__be16 result;
+	__be16 req_resp;
+} __attribute__((packed));
+
+#define RPMB_PROGRAM_KEY       0x1    /* Program RPMB Authentication Key */
+#define RPMB_GET_WRITE_COUNTER 0x2    /* Read RPMB write counter */
+#define RPMB_WRITE_DATA        0x3    /* Write data to RPMB partition */
+#define RPMB_READ_DATA         0x4    /* Read data from RPMB partition */
+#define RPMB_RESULT_READ       0x5    /* Read result request  (Internal) */
+
+#define RPMB_REQ2RESP(_OP) ((_OP) << 8)
+#define RPMB_RESP2REQ(_OP) ((_OP) >> 8)
+
+/* length of the part of the frame used for HMAC computation */
+#define hmac_data_len \
+	(sizeof(struct rpmb_frame) - offsetof(struct rpmb_frame, data))
+
+/**
+ * enum rpmb_op_result - rpmb operation results
+ *
+ * @RPMB_ERR_OK:       operation successful
+ * @RPMB_ERR_GENERAL:  general failure
+ * @RPMB_ERR_AUTH:     mac doesn't match or ac calculation failure
+ * @RPMB_ERR_COUNTER:  counter doesn't match or counter increment failure
+ * @RPMB_ERR_ADDRESS:  address out of range or wrong address alignment
+ * @RPMB_ERR_WRITE:    data, counter, or result write failure
+ * @RPMB_ERR_READ:     data, counter, or result read failure
+ * @RPMB_ERR_NO_KEY:   authentication key not yet programmed
+ *
+ * @RPMB_ERR_COUNTER_EXPIRED:  counter expired
+ */
+enum rpmb_op_result {
+	RPMB_ERR_OK      = 0x0000,
+	RPMB_ERR_GENERAL = 0x0001,
+	RPMB_ERR_AUTH    = 0x0002,
+	RPMB_ERR_COUNTER = 0x0003,
+	RPMB_ERR_ADDRESS = 0x0004,
+	RPMB_ERR_WRITE   = 0x0005,
+	RPMB_ERR_READ    = 0x0006,
+	RPMB_ERR_NO_KEY  = 0x0007,
+
+	RPMB_ERR_COUNTER_EXPIRED = 0x0080
+};
+
+/**
+ * struct rpmb_ioc_cmd - rpmb operation request command
+ *
+ * @req:                request type:  must match the in frame req_resp
+ *                          program key
+ *                          get write counter
+ *                          write/read data
+ * @in_frames_count:    number of in frames
+ * @out_frames_count:   number of out frames
+ * @in_frames_ptr:      a pointer to the input frames buffer
+ * @out_frames_ptr:     a pointer to output frames buffer
+ */
+struct rpmb_ioc_cmd {
+	__u32  req;
+	__u32  in_frames_count;
+	__u32  out_frames_count;
+	__aligned_u64 in_frames_ptr;
+	__aligned_u64 out_frames_ptr;
+};
+
+#define rpmb_ioc_cmd_set_in_frames(_ic, _ptr) \
+	(_ic).in_frames_ptr = (__aligned_u64)(intptr_t)(_ptr)
+#define rpmb_ioc_cmd_set_out_frames(_ic, _ptr) \
+	(_ic).out_frames_ptr = (__aligned_u64)(intptr_t)(_ptr)
+
+#define RPMB_IOC_REQ _IOWR(0xB5, 1, struct rpmb_ioc_cmd)
+
+#endif /* _UAPI_LINUX_RPMB_H_ */
-- 
2.5.5

[PATCH v4 5/8] char: rpmb: add RPMB simulation device

[Tomas Winkler]

This is a simple platform device used for testing
the rpmb subsystem.

The module currently supports two configuration options:
1. max_wr_blks: for specifying max blocks that can be written
in a single command
2. daunits:  used to set storage capacity in 128K units.

V2: remove .owner setting, it is set automatically
V3: 1. Add shutdown handler (similar to ufshcd)
    2. Commit message fix
V4: use select RPMB to ensure valid configuration

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
 drivers/char/rpmb/Kconfig    |  10 +
 drivers/char/rpmb/Makefile   |   1 +
 drivers/char/rpmb/rpmb_sim.c | 589 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 600 insertions(+)
 create mode 100644 drivers/char/rpmb/rpmb_sim.c

diff --git a/drivers/char/rpmb/Kconfig b/drivers/char/rpmb/Kconfig
index 6794be9fcc5e..c21b3934249f 100644
--- a/drivers/char/rpmb/Kconfig
+++ b/drivers/char/rpmb/Kconfig
@@ -13,3 +13,13 @@ config RPMB_INTF_DEV
 	help
 	  Say yes here if you want to access RPMB from user space
 	  via character device interface /dev/rpmb%d
+
+
+config RPMB_SIM
+	tristate "RPMB partition device simulator"
+	default n
+	select RPMB
+	select CRYPTO_SHA256
+	select CRYPTO_HMAC
+	help
+	  RPMB partition simulator used for testing the RPMB subsystem
diff --git a/drivers/char/rpmb/Makefile b/drivers/char/rpmb/Makefile
index b5dc087b1299..81f924fd9a87 100644
--- a/drivers/char/rpmb/Makefile
+++ b/drivers/char/rpmb/Makefile
@@ -1,5 +1,6 @@
 obj-$(CONFIG_RPMB) += rpmb.o
 rpmb-objs += core.o
 rpmb-$(CONFIG_RPMB_INTF_DEV) += cdev.o
+obj-$(CONFIG_RPMB_SIM) += rpmb_sim.o
 
 ccflags-y += -D__CHECK_ENDIAN__
diff --git a/drivers/char/rpmb/rpmb_sim.c b/drivers/char/rpmb/rpmb_sim.c
new file mode 100644
index 000000000000..f2565f406cc1
--- /dev/null
+++ b/drivers/char/rpmb/rpmb_sim.c
@@ -0,0 +1,589 @@
+/******************************************************************************
+ * This file is provided under a dual BSD/GPLv2 license.  When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * Copyright(c) 2016 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * The full GNU General Public License is included in this distribution
+ * in the file called LICENSE.GPL.
+ *
+ * Contact Information:
+ *	Intel Corporation.
+ *	linux-mei@linux.intel.com
+ *	http://www.intel.com
+ *
+ * BSD LICENSE
+ *
+ * Copyright(c) 2016 Intel Corporation. All rights reserved.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *  * Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ *  * Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in
+ *    the documentation and/or other materials provided with the
+ *    distribution.
+ *  * Neither the name Intel Corporation nor the names of its
+ *    contributors may be used to endorse or promote products derived
+ *    from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *****************************************************************************/
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/platform_device.h>
+#include <linux/device.h>
+#include <crypto/hash.h>
+#include <linux/scatterlist.h>
+#include <linux/sizes.h>
+
+#include <linux/rpmb.h>
+
+static const char id[] = "RPMB:SIM";
+#define CAPACITY_UNIT SZ_128K
+#define CAPACITY_MIN  SZ_128K
+#define CAPACITY_MAX  SZ_16M
+#define BLK_UNIT      SZ_256
+
+static unsigned int max_wr_blks = 1;
+module_param(max_wr_blks, uint, 0644);
+MODULE_PARM_DESC(max_wr_blks,
+	"max blocks that can be written in a single command (default: 1)");
+
+static unsigned int daunits = 1;
+module_param(daunits, uint, 0644);
+MODULE_PARM_DESC(daunits,
+	"number of data area units of 128K (default: 1)");
+
+struct blk {
+	u8 data[BLK_UNIT];
+};
+
+/**
+ * struct rpmb_sim_dev
+ *
+ * @dev:  back pointer to the platform device
+ * @rdev: rpmb device
+ * @auth_key: Authentication key register which is used to authenticate
+ *            accesses when MAC is calculated;
+ * @auth_key_set: true if auth key was set
+ * @write_counter: Counter value for the total amount of successful
+ *             authenticated data write requests made by the host.
+ *             The initial value of this register after production is 00000000h.
+ *             The value will be incremented by one along with each successful
+ *             programming access. The value cannot be reset. After the counter
+ *             has reached the maximum value of FFFFFFFFh,
+ *             it will not be incremented anymore (overflow prevention)
+ * @hash_tfm:  hmac(sha256) tfm
+ *
+ * @capacity: size of the partition in bytes multiple of 128K
+ * @blkcnt:   block count
+ * @da:       data area in blocks
+ */
+struct rpmb_sim_dev {
+	struct device *dev;
+	struct rpmb_dev *rdev;
+	u8 auth_key[32];
+	bool auth_key_set;
+	u32 write_counter;
+	struct crypto_shash *hash_tfm;
+
+	size_t capacity;
+	size_t blkcnt;
+	struct blk *da;
+};
+
+static __be16 op_result(struct rpmb_sim_dev *rsdev, u16 result)
+{
+	if (!rsdev->auth_key_set)
+		return cpu_to_be16(RPMB_ERR_NO_KEY);
+
+	if (rsdev->write_counter == 0xFFFFFFFF)
+		result |=  RPMB_ERR_COUNTER_EXPIRED;
+
+	return cpu_to_be16(result);
+}
+
+static __be16 req_to_resp(u16 req)
+{
+	return cpu_to_be16(RPMB_REQ2RESP(req));
+}
+
+static int rpmb_sim_calc_hmac(struct rpmb_sim_dev *rsdev,
+			      struct rpmb_frame *frames,
+			      unsigned int blks, u8 *mac)
+{
+	SHASH_DESC_ON_STACK(desc, rsdev->hash_tfm);
+	int i;
+	int ret;
+
+	desc->tfm = rsdev->hash_tfm;
+	desc->flags = 0;
+
+	ret = crypto_shash_init(desc);
+	if (ret)
+		goto out;
+
+	for (i = 0; i < blks; i++) {
+		ret = crypto_shash_update(desc, frames[i].data, hmac_data_len);
+		if (ret)
+			goto out;
+	}
+	ret = crypto_shash_final(desc, mac);
+out:
+	if (ret)
+		dev_err(rsdev->dev, "digest error = %d", ret);
+
+	return ret;
+}
+
+static int rpmb_op_not_programmed(struct rpmb_sim_dev *rsdev,
+				  struct rpmb_data *rpmbd)
+{
+	struct rpmb_frame *in_frame, *out_frame;
+	u16 req;
+
+	in_frame = &rpmbd->in_frames[0];
+	req = be16_to_cpu(in_frame->req_resp);
+
+	out_frame = &rpmbd->out_frames[0];
+
+	out_frame->req_resp = req_to_resp(req);
+	out_frame->result = op_result(rsdev, RPMB_ERR_NO_KEY);
+
+	dev_err(rsdev->dev, "not programmed\n");
+
+	return 0;
+}
+
+static int rpmb_op_program_key(struct rpmb_sim_dev *rsdev,
+			       struct rpmb_data *rpmbd)
+{
+	struct rpmb_frame *in_frame, *out_frame;
+	u16 req;
+	int ret;
+	u16 err = RPMB_ERR_OK;
+
+	in_frame = rpmbd->in_frames;
+	req = be16_to_cpu(in_frame[0].req_resp);
+
+	if (req != RPMB_PROGRAM_KEY)
+		return -EINVAL;
+
+	if (rsdev->auth_key_set) {
+		dev_err(rsdev->dev, "key allread set\n");
+		err = RPMB_ERR_WRITE;
+		goto out;
+	}
+
+	ret = crypto_shash_setkey(rsdev->hash_tfm, in_frame[0].key_mac, 32);
+	if (ret) {
+		dev_err(rsdev->dev, "set key failed = %d\n", ret);
+		err = RPMB_ERR_GENERAL;
+		goto out;
+	}
+
+	dev_dbg(rsdev->dev, "digest size %u",
+		crypto_shash_digestsize(rsdev->hash_tfm));
+
+	memcpy(rsdev->auth_key, in_frame[0].key_mac, 32);
+	rsdev->auth_key_set = true;
+out:
+	out_frame = rpmbd->out_frames;
+	memset(out_frame, 0, sizeof(struct rpmb_frame));
+
+	out_frame[0].req_resp = req_to_resp(req);
+	out_frame[0].result = op_result(rsdev, err);
+
+	return 0;
+}
+
+
+static int rpmb_op_get_wr_conter(struct rpmb_sim_dev *rsdev,
+				 struct rpmb_data *rpmbd)
+{
+	struct rpmb_frame *in_frame, *out_frame;
+	u16 req;
+	u16 err;
+
+	in_frame = rpmbd->in_frames;
+	req = be16_to_cpu(in_frame[0].req_resp);
+
+	if (req != RPMB_GET_WRITE_COUNTER)
+		return -EINVAL;
+
+	out_frame = rpmbd->out_frames;
+	memset(out_frame, 0, sizeof(struct rpmb_frame));
+
+	out_frame[0].req_resp = req_to_resp(req);
+	out_frame[0].write_counter = cpu_to_be32(rsdev->write_counter);
+	memcpy(out_frame[0].nonce, in_frame[0].nonce, 16);
+
+	err = RPMB_ERR_OK;
+	out_frame[0].result = op_result(rsdev, err);
+
+	if (rpmb_sim_calc_hmac(rsdev, out_frame, 1, out_frame->key_mac))
+		err = RPMB_ERR_READ;
+
+	out_frame[0].result = op_result(rsdev, err);
+
+	return 0;
+}
+
+static int rpmb_op_write_data(struct rpmb_sim_dev *rsdev,
+			      struct rpmb_data *rpmbd)
+{
+	struct rpmb_frame *in_frame, *out_frame;
+	u8 mac[32];
+	u16 req, err, addr, blks;
+	unsigned int i;
+	int ret = 0;
+
+	in_frame = rpmbd->in_frames;
+	req = be16_to_cpu(in_frame[0].req_resp);
+
+	if (req != RPMB_WRITE_DATA)
+		return -EINVAL;
+
+
+	if (rsdev->write_counter == 0xFFFFFFFF) {
+		err = RPMB_ERR_WRITE;
+		goto out;
+	}
+
+	blks = be16_to_cpu(in_frame[0].block_count);
+	if (blks == 0 || blks > rpmbd->in_frames_cnt) {
+		ret = -EINVAL;
+		err = RPMB_ERR_GENERAL;
+		goto out;
+	}
+
+	if (blks > max_wr_blks) {
+		err = RPMB_ERR_WRITE;
+		goto out;
+	}
+
+
+	addr = be16_to_cpu(in_frame[0].addr);
+	if (addr >= rsdev->blkcnt) {
+		err = RPMB_ERR_ADDRESS;
+		goto out;
+	}
+
+	if (rpmb_sim_calc_hmac(rsdev, in_frame, blks, mac)) {
+		err = RPMB_ERR_AUTH;
+		goto out;
+	}
+
+	/* mac is in the last frame */
+	if (memcmp(mac, in_frame[blks - 1].key_mac, sizeof(mac)) != 0) {
+		err = RPMB_ERR_AUTH;
+		goto out;
+	}
+
+	if (be32_to_cpu(in_frame[0].write_counter) != rsdev->write_counter) {
+		err = RPMB_ERR_COUNTER;
+		goto out;
+	}
+
+	if (addr + blks > rsdev->blkcnt) {
+		err = RPMB_ERR_WRITE;
+		goto out;
+	}
+
+	err = RPMB_ERR_OK;
+	for (i = 0; i < blks; i++)
+		memcpy(rsdev->da[addr + i].data, in_frame[i].data, BLK_UNIT);
+
+	rsdev->write_counter++;
+
+out:
+	out_frame = &rpmbd->out_frames[0];
+	memset(out_frame, 0, sizeof(struct rpmb_frame));
+
+	if (err == RPMB_ERR_OK) {
+		out_frame[0].write_counter = cpu_to_be32(rsdev->write_counter);
+		memcpy(out_frame[0].key_mac, mac, sizeof(mac));
+	}
+	out_frame[0].req_resp = req_to_resp(req);
+	out_frame[0].result = op_result(rsdev, err);
+
+	return ret;
+}
+
+static int rpmb_op_read_data(struct rpmb_sim_dev *rsdev,
+			     struct rpmb_data *rpmbd)
+{
+	struct rpmb_frame *in_frame, *out_frame;
+	u8 mac[32];
+	u16 req, err, addr, blks;
+	unsigned int i;
+	int ret;
+
+	in_frame = rpmbd->in_frames;
+	req = be16_to_cpu(in_frame[0].req_resp);
+	if (req != RPMB_READ_DATA)
+		return -EINVAL;
+
+
+	out_frame = rpmbd->out_frames;
+
+	blks = be16_to_cpu(in_frame[0].block_count);
+	if (blks == 0 || blks > rpmbd->out_frames_cnt) {
+		ret = -EINVAL;
+		memset(out_frame, 0, sizeof(struct rpmb_frame));
+		err = RPMB_ERR_READ;
+		goto out;
+	}
+
+	ret = 0;
+	memset(out_frame, 0, sizeof(struct rpmb_frame) * blks);
+
+	addr = be16_to_cpu(in_frame[0].addr);
+	if (addr >= rsdev->blkcnt) {
+		err = RPMB_ERR_ADDRESS;
+		goto out;
+	}
+
+	if (addr + blks > rsdev->blkcnt) {
+		err = RPMB_ERR_READ;
+		goto out;
+	}
+
+	for (i = 0; i < blks; i++) {
+		memcpy(out_frame[i].data, rsdev->da[addr + i].data, BLK_UNIT);
+		memcpy(out_frame[i].nonce, in_frame[0].nonce, 16);
+		out_frame[i].req_resp = req_to_resp(req);
+		out_frame[i].addr = in_frame[0].addr;
+		out_frame[i].block_count = cpu_to_be16(blks);
+	}
+
+	if (rpmb_sim_calc_hmac(rsdev, out_frame, blks, mac)) {
+		err = RPMB_ERR_AUTH;
+		goto out;
+	}
+
+	memcpy(out_frame[blks - 1].key_mac, mac, sizeof(mac));
+
+	err = RPMB_ERR_OK;
+out:
+
+	/* FIXME: not sure if this has to be updated in each frame */
+	out_frame[0].result = op_result(rsdev, err);
+
+
+	return ret;
+}
+
+static int rpmb_sim_sequence(struct rpmb_sim_dev *rsdev,
+			     struct rpmb_data *rpmbd)
+{
+	u16 type;
+	int ret;
+
+	type = rpmbd->req_type;
+
+	if (rpmbd->out_frames == NULL || rpmbd->in_frames == NULL ||
+	    rpmbd->out_frames_cnt == 0 || rpmbd->in_frames_cnt == 0)
+		return -EINVAL;
+
+	if (rsdev->auth_key_set == false && type != RPMB_PROGRAM_KEY)
+		return rpmb_op_not_programmed(rsdev, rpmbd);
+
+	switch (type) {
+	case RPMB_PROGRAM_KEY:
+		ret = rpmb_op_program_key(rsdev, rpmbd);
+		break;
+	case RPMB_WRITE_DATA:
+		ret = rpmb_op_write_data(rsdev, rpmbd);
+		break;
+	case RPMB_GET_WRITE_COUNTER:
+		ret = rpmb_op_get_wr_conter(rsdev, rpmbd);
+		break;
+	case RPMB_READ_DATA:
+		ret = rpmb_op_read_data(rsdev, rpmbd);
+		break;
+	default:
+		ret = -EINVAL;
+		goto out;
+	}
+out:
+	return ret;
+}
+
+static int rpmb_sim_send_rpmb_req(struct device *dev, struct rpmb_data *rpmbd)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct rpmb_sim_dev *rsdev = platform_get_drvdata(pdev);
+
+	if (!rsdev)
+		return -EINVAL;
+
+	if (!rpmbd)
+		return -EINVAL;
+
+	return rpmb_sim_sequence(rsdev, rpmbd);
+}
+
+static struct rpmb_ops rpmb_sim_ops = {
+	.send_rpmb_req = rpmb_sim_send_rpmb_req,
+	.type = RPMB_TYPE_EMMC,
+};
+
+static int rpmb_sim_hmac_256_alloc(struct rpmb_sim_dev *rsdev)
+{
+	struct crypto_shash *hash_tfm;
+
+	hash_tfm = crypto_alloc_shash("hmac(sha256)", 0, 0);
+	if (IS_ERR(hash_tfm))
+		return PTR_ERR(hash_tfm);
+
+	rsdev->hash_tfm = hash_tfm;
+
+	dev_dbg(rsdev->dev, "hamac(sha256) registered\n");
+	return 0;
+}
+
+static void rpmb_sim_hmac_256_free(struct rpmb_sim_dev *rsdev)
+{
+	if (rsdev->hash_tfm)
+		crypto_free_shash(rsdev->hash_tfm);
+
+	rsdev->hash_tfm = NULL;
+}
+
+static int rpmb_sim_probe(struct platform_device *pdev)
+{
+	struct rpmb_sim_dev *rsdev;
+	int ret;
+
+	rsdev = kzalloc(sizeof(struct rpmb_sim_dev), GFP_KERNEL);
+	if (!rsdev)
+		return -ENOMEM;
+
+	rsdev->dev = &pdev->dev;
+
+	ret = rpmb_sim_hmac_256_alloc(rsdev);
+	if (ret)
+		goto err;
+
+	rsdev->capacity = CAPACITY_UNIT * daunits;
+	rsdev->blkcnt  = rsdev->capacity / BLK_UNIT;
+	rsdev->da = kzalloc(rsdev->capacity, GFP_KERNEL);
+	if (!rsdev->da) {
+		ret = -ENOMEM;
+		goto err;
+	}
+
+	rpmb_sim_ops.dev_id_len = strlen(id);
+	rpmb_sim_ops.dev_id = id;
+	rpmb_sim_ops.reliable_wr_cnt = max_wr_blks;
+
+	rsdev->rdev = rpmb_dev_register(rsdev->dev, &rpmb_sim_ops);
+	if (IS_ERR(rsdev->rdev)) {
+		ret = PTR_ERR(rsdev->rdev);
+		goto err;
+	}
+
+	dev_info(&pdev->dev, "registered RPMB capacity = %zu of %zu blocks\n",
+		 rsdev->capacity, rsdev->blkcnt);
+
+	platform_set_drvdata(pdev, rsdev);
+
+	return 0;
+err:
+	rpmb_sim_hmac_256_free(rsdev);
+	if (rsdev)
+		kfree(rsdev->da);
+	kfree(rsdev);
+	return ret;
+}
+
+static int rpmb_sim_remove(struct platform_device *pdev)
+{
+	struct rpmb_sim_dev *rsdev;
+
+	rsdev = platform_get_drvdata(pdev);
+
+	rpmb_dev_unregister(rsdev->dev);
+
+	platform_set_drvdata(pdev, NULL);
+
+	rpmb_sim_hmac_256_free(rsdev);
+
+	kfree(rsdev->da);
+	kfree(rsdev);
+	return 0;
+}
+
+void rpmb_sim_shutdown(struct platform_device *pdev)
+{
+	rpmb_sim_remove(pdev);
+}
+
+static struct platform_driver rpmb_sim_driver = {
+	.driver = {
+		.name  = "rpmb_sim",
+	},
+	.probe         = rpmb_sim_probe,
+	.remove        = rpmb_sim_remove,
+	.shutdown      = rpmb_sim_shutdown,
+};
+
+static struct platform_device *rpmb_sim_pdev;
+
+static int __init rpmb_sim_init(void)
+{
+	int ret;
+
+	rpmb_sim_pdev = platform_device_register_simple("rpmb_sim", -1,
+							NULL, 0);
+
+	if (IS_ERR(rpmb_sim_pdev))
+		return PTR_ERR(rpmb_sim_pdev);
+
+	ret = platform_driver_register(&rpmb_sim_driver);
+	if (ret)
+		platform_device_unregister(rpmb_sim_pdev);
+
+	return ret;
+}
+
+static void __exit rpmb_sim_exit(void)
+{
+	platform_device_unregister(rpmb_sim_pdev);
+	platform_driver_unregister(&rpmb_sim_driver);
+}
+
+module_init(rpmb_sim_init);
+module_exit(rpmb_sim_exit);
+
+MODULE_AUTHOR("Tomas Winkler <tomas.winkler@intel.com");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_ALIAS("platform:rpmb_sim");
-- 
2.5.5

[PATCH v4 6/8] tools rpmb: add RPBM access tool

[Tomas Winkler]

Add simple RPMB host testing tool. It can be used
to program key, write and read data block, and retrieve
write counter.

V2: resend
V3: fix missing objtool
V4: add verbose option

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
 MAINTAINERS           |   1 +
 tools/Makefile        |  14 +-
 tools/rpmb/.gitignore |   2 +
 tools/rpmb/Makefile   |  32 ++
 tools/rpmb/rpmb.c     | 862 ++++++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 906 insertions(+), 5 deletions(-)
 create mode 100644 tools/rpmb/.gitignore
 create mode 100644 tools/rpmb/Makefile
 create mode 100644 tools/rpmb/rpmb.c

diff --git a/MAINTAINERS b/MAINTAINERS
index bf806f2d3e19..41db922aaf38 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9699,6 +9699,7 @@ F:	drivers/char/rpmb/*
 F:	include/uapi/linux/rpmb.h
 F:	include/linux/rpmb.h
 F:	Documentation/ABI/testing/sysfs-class-rpmb
+F:	tools/rpmb/
 
 RTL2830 MEDIA DRIVER
 M:	Antti Palosaari <crope@iki.fi>
diff --git a/tools/Makefile b/tools/Makefile
index f10b64d8c674..b90003945af1 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -20,6 +20,7 @@ help:
 	@echo '  lguest                 - a minimal 32-bit x86 hypervisor'
 	@echo '  net                    - misc networking tools'
 	@echo '  perf                   - Linux performance measurement and analysis tool'
+	@echo '  rpmb                   - Replay protected memory block access tool'
 	@echo '  selftests              - various kernel selftests'
 	@echo '  spi                    - spi tools'
 	@echo '  objtool                - an ELF object analysis tool'
@@ -56,7 +57,7 @@ acpi: FORCE
 cpupower: FORCE
 	$(call descend,power/$@)
 
-cgroup firewire hv guest spi usb virtio vm net iio gpio objtool: FORCE
+cgroup firewire hv guest rpmb spi usb virtio vm net iio gpio objtool: FORCE
 	$(call descend,$@)
 
 liblockdep: FORCE
@@ -86,7 +87,7 @@ freefall: FORCE
 	$(call descend,laptop/$@)
 
 all: acpi cgroup cpupower hv firewire lguest \
-		perf selftests turbostat usb \
+		perf rpmb selftests turbostat usb \
 		virtio vm net x86_energy_perf_policy \
 		tmon freefall objtool
 
@@ -96,7 +97,7 @@ acpi_install:
 cpupower_install:
 	$(call descend,power/$(@:_install=),install)
 
-cgroup_install firewire_install hv_install lguest_install perf_install usb_install virtio_install vm_install net_install objtool_install:
+cgroup_install firewire_install hv_install lguest_install perf_install rpmb_install usb_install virtio_install vm_install net_install objtool_install:
 	$(call descend,$(@:_install=),install)
 
 selftests_install:
@@ -115,7 +116,7 @@ kvm_stat_install:
 	$(call descend,kvm/$(@:_install=),install)
 
 install: acpi_install cgroup_install cpupower_install hv_install firewire_install lguest_install \
-		perf_install selftests_install turbostat_install usb_install \
+		perf_install rpmb_install selftests_install turbostat_install usb_install \
 		virtio_install vm_install net_install x86_energy_perf_policy_install \
 		tmon_install freefall_install objtool_install kvm_stat_install
 
@@ -144,6 +145,9 @@ perf_clean:
 	$(Q)mkdir -p $(PERF_O) .
 	$(Q)$(MAKE) --no-print-directory -C perf O=$(PERF_O) subdir= clean
 
+rpmb_clean:
+	$(call descend,$(@:_clean=),clean)
+
 selftests_clean:
 	$(call descend,testing/$(@:_clean=),clean)
 
@@ -160,7 +164,7 @@ build_clean:
 	$(call descend,build,clean)
 
 clean: acpi_clean cgroup_clean cpupower_clean hv_clean firewire_clean lguest_clean \
-		perf_clean selftests_clean turbostat_clean spi_clean usb_clean virtio_clean \
+		perf_clean rpmb_clean selftests_clean turbostat_clean spi_clean usb_clean virtio_clean \
 		vm_clean net_clean iio_clean x86_energy_perf_policy_clean tmon_clean \
 		freefall_clean build_clean libbpf_clean libsubcmd_clean liblockdep_clean \
 		gpio_clean objtool_clean
diff --git a/tools/rpmb/.gitignore b/tools/rpmb/.gitignore
new file mode 100644
index 000000000000..218f680548e6
--- /dev/null
+++ b/tools/rpmb/.gitignore
@@ -0,0 +1,2 @@
+*.o
+rpmb
diff --git a/tools/rpmb/Makefile b/tools/rpmb/Makefile
new file mode 100644
index 000000000000..e0e51f306dac
--- /dev/null
+++ b/tools/rpmb/Makefile
@@ -0,0 +1,32 @@
+CC = $(CROSS_COMPILE)gcc
+LD = $(CROSS_COMPILE)ld
+PKG_CONFIG = $(CROSS_COMPILE)pkg-config
+
+ifeq ($(srctree),)
+srctree := $(patsubst %/,%,$(dir $(shell pwd)))
+srctree := $(patsubst %/,%,$(dir $(srctree)))
+#$(info Determined 'srctree' to be $(srctree))
+endif
+
+INSTALL = install
+prefix ?= /usr/local
+bindir = $(prefix)/bin
+
+
+CFLAGS += -D__EXPORTED_HEADERS__
+CFLAGS += -I$(srctree)/include/uapi -I$(srctree)/include
+LDLIBS += $(shell $(PKG_CONFIG) --libs libcrypto)
+
+prog := rpmb
+
+all : $(prog)
+
+$(prog): rpmb.o
+
+clean :
+	$(RM) $(prog) *.o
+
+install: $(prog)
+	$(INSTALL) -m755 -d $(DESTDIR)$(bindir)
+	$(INSTALL) $(prog) $(DESTDIR)$(bindir)
+
diff --git a/tools/rpmb/rpmb.c b/tools/rpmb/rpmb.c
new file mode 100644
index 000000000000..99b987fd0c01
--- /dev/null
+++ b/tools/rpmb/rpmb.c
@@ -0,0 +1,862 @@
+/******************************************************************************
+ * This file is provided under a dual BSD/GPLv2 license.  When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * Copyright(c) 2016 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * The full GNU General Public License is included in this distribution
+ * in the file called LICENSE.GPL.
+ *
+ * Contact Information:
+ *	Intel Corporation.
+ *	linux-mei@linux.intel.com
+ *	http://www.intel.com
+ *
+ * BSD LICENSE
+ *
+ * Copyright(c) 2016 Intel Corporation. All rights reserved.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *  * Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ *  * Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in
+ *    the documentation and/or other materials provided with the
+ *    distribution.
+ *  * Neither the name Intel Corporation nor the names of its
+ *    contributors may be used to endorse or promote products derived
+ *    from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *****************************************************************************/
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <dirent.h>
+#include <sys/stat.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <libgen.h>
+#include <limits.h>
+#include <ctype.h>
+#include <errno.h>
+#include <stdint.h>
+#include <stdbool.h>
+
+#include <openssl/engine.h>
+#include <openssl/hmac.h>
+
+#include "linux/rpmb.h"
+
+#define RPMB_KEY_SIZE 32
+#define RPMB_MAC_SIZE 32
+#define RPMB_NONCE_SIZE 16
+
+bool verbose;
+#define rpmb_dbg(fmt, ARGS...) do {                     \
+	if (verbose)                                    \
+		fprintf(stderr, "rpmb: " fmt, ##ARGS);  \
+} while (0)
+
+#define rpmb_msg(fmt, ARGS...) \
+	fprintf(stderr, "rpmb: " fmt, ##ARGS)
+
+#define rpmb_err(fmt, ARGS...) \
+	fprintf(stderr, "rpmb: error: " fmt, ##ARGS)
+
+static const char *rpmb_op_str(uint16_t op)
+{
+#define RPMB_OP(_op) case RPMB_##_op: return #_op
+
+	switch (op) {
+	RPMB_OP(PROGRAM_KEY);
+	RPMB_OP(GET_WRITE_COUNTER);
+	RPMB_OP(WRITE_DATA);
+	RPMB_OP(READ_DATA);
+	RPMB_OP(RESULT_READ);
+	break;
+	default:
+		return "unknown";
+	}
+}
+
+static const char *rpmb_result_str(enum rpmb_op_result result)
+{
+#define str(x) #x
+#define RPMB_ERR(_res) case RPMB_ERR_##_res:         \
+	{ if (result & RPMB_ERR_COUNTER_EXPIRED)     \
+		return "COUNTER_EXPIRE:" str(_res);  \
+	else                                         \
+		return #_res;                        \
+	}
+
+	switch (result & 0x000F) {
+	RPMB_ERR(OK);
+	RPMB_ERR(GENERAL);
+	RPMB_ERR(AUTH);
+	RPMB_ERR(COUNTER);
+	RPMB_ERR(ADDRESS);
+	RPMB_ERR(WRITE);
+	RPMB_ERR(READ);
+	RPMB_ERR(NO_KEY);
+	break;
+	default:
+		return "unknown";
+	}
+#undef RPMB_ERR
+};
+
+static inline void __dump_buffer(const char *buf)
+{
+	fprintf(stderr, "%s\n", buf);
+}
+
+static void
+dump_hex_buffer(const char *title, const unsigned char *buf, size_t len)
+{
+	const size_t pbufsz = 16 * 3;
+	char pbuf[pbufsz];
+	int j = 0;
+
+	if (title)
+		fprintf(stderr, "%s\n", title);
+	while (len-- > 0) {
+		snprintf(&pbuf[j], pbufsz - j, "%02X ", *buf++);
+		j += 3;
+		if (j == 16 * 3) {
+			__dump_buffer(pbuf);
+			j = 0;
+		}
+	}
+	if (j)
+		__dump_buffer(pbuf);
+}
+
+static void dbg_dump_frame(const char *title, const struct rpmb_frame *f)
+{
+	uint16_t result, req_resp;
+
+	if (!verbose)
+		return;
+
+	if (title)
+		fprintf(stderr, "%s\n", title);
+
+	result = be16toh(f->result);
+	req_resp = be16toh(f->req_resp);
+	if (req_resp & 0xf00)
+		req_resp = RPMB_RESP2REQ(req_resp);
+
+	dump_hex_buffer("key_mac: ", f->key_mac, 32);
+	dump_hex_buffer("data: ", f->data, 256);
+	dump_hex_buffer("nonce: ", f->nonce, 16);
+	fprintf(stderr, "write_counter: %u\n", be32toh(f->write_counter));
+	fprintf(stderr, "address:  %0X\n", be16toh(f->addr));
+	fprintf(stderr, "block_count: %u\n", be16toh(f->block_count));
+	fprintf(stderr, "result %s:%d\n", rpmb_result_str(result), result);
+	fprintf(stderr, "req_resp %s\n", rpmb_op_str(req_resp));
+}
+
+static int open_dev_file(const char *devfile)
+{
+	int fd;
+
+	fd = open(devfile, O_RDWR);
+	if (fd < 0)
+		rpmb_err("Cannot open: %s: %s\n", devfile, strerror(errno));
+	return fd;
+}
+
+static int open_rd_file(const char *datafile, const char *type)
+{
+	int fd;
+
+	if (!strcmp(datafile, "-"))
+		fd = STDIN_FILENO;
+	else
+		fd = open(datafile, O_RDONLY);
+
+	if (fd < 0)
+		rpmb_err("Cannot open %s: %s: %s\n",
+			type, datafile, strerror(errno));
+
+	return fd;
+}
+
+static int open_wr_file(const char *datafile, const char *type)
+{
+	int fd;
+
+	if (!strcmp(datafile, "-"))
+		fd = STDOUT_FILENO;
+	else
+		fd = open(datafile, O_WRONLY | O_CREAT | O_APPEND,
+					   S_IRUSR | S_IWUSR);
+	if (fd < 0)
+		rpmb_err("Cannot open %s: %s: %s\n",
+			type, datafile, strerror(errno));
+	return fd;
+}
+
+static void close_fd(int fd)
+{
+	if (fd > 0 && fd != STDIN_FILENO && fd != STDOUT_FILENO)
+		close(fd);
+}
+
+/* blocking rw wrapper */
+#define rw(func, fd, buf, size)                              \
+({                                                           \
+	ssize_t nread = 0, n;                                \
+	char *_buf = (char *)buf;                            \
+	do {                                                 \
+		n = func(fd, _buf + nread, size - nread);    \
+		if (n == -1 && errno != EINTR) {             \
+			nread = -1;                          \
+			break;                               \
+		} else if (n > 0) {                          \
+			nread += n;                          \
+		}                                            \
+	} while (n != 0 && (size_t)nread != size);           \
+	nread;                                               \
+})
+
+
+static ssize_t read_file(int fd, unsigned char *data, size_t size)
+{
+	ssize_t ret;
+
+	ret = rw(read, fd, data, size);
+	if (ret < 0) {
+		rpmb_err("cannot read file: %s\n", strerror(errno));
+	} else if (ret != size) {
+		rpmb_err("read %zd but must be %zu bytes length\n",
+			ret, size);
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+static ssize_t write_file(int fd, unsigned char *data, size_t size)
+{
+	ssize_t ret;
+
+	ret = rw(write, fd, data, size);
+	if (ret < 0) {
+		rpmb_err("cannot read file: %s\n", strerror(errno));
+	} else if (ret != size) {
+		rpmb_err("data is %zd but must be %zu bytes length\n",
+			ret, size);
+		ret = -EINVAL;
+	}
+	return ret;
+}
+
+static int rpmb_calc_hmac_sha256(struct rpmb_frame *frames, size_t blocks_cnt,
+				 const unsigned char key[],
+				 unsigned int key_size,
+				 unsigned char mac[],
+				 unsigned int mac_size)
+{
+	HMAC_CTX ctx;
+	int ret;
+	int i;
+
+	 /* SSL returns 1 on success 0 on failure */
+
+	HMAC_CTX_init(&ctx);
+	ret = HMAC_Init_ex(&ctx, key, key_size, EVP_sha256(), NULL);
+	if (ret == 0)
+		goto out;
+	for (i = 0; i < blocks_cnt; i++)
+		HMAC_Update(&ctx, frames[i].data, hmac_data_len);
+
+	ret = HMAC_Final(&ctx, mac, &mac_size);
+	if (ret == 0)
+		goto out;
+	if (mac_size != RPMB_MAC_SIZE)
+		ret = 0;
+
+	ret = 1;
+out:
+	HMAC_CTX_cleanup(&ctx);
+	return ret == 1 ? 0 : -1;
+}
+
+static int rpmb_ioctl(int fd, uint16_t req,
+	const struct rpmb_frame *frames_in,
+	unsigned int in_cnt,
+	struct rpmb_frame *frames_out,
+	unsigned int out_cnt)
+{
+	int ret;
+	struct rpmb_ioc_cmd icmd;
+
+	icmd.req = req;
+	icmd.in_frames_count = in_cnt;
+	rpmb_ioc_cmd_set_in_frames(icmd, frames_in);
+	rpmb_ioc_cmd_set_out_frames(icmd, frames_out);
+	icmd.out_frames_count = out_cnt;
+
+	rpmb_dbg("RPMB OP: %s\n", rpmb_op_str(req));
+	dbg_dump_frame("In Frame: ", frames_in);
+	ret = ioctl(fd, RPMB_IOC_REQ, &icmd);
+	dbg_dump_frame("Out Frame: ", frames_out);
+	if (ret < 0)
+		rpmb_err("ioctl failure %d: %s\n", ret, strerror(errno));
+
+	return ret;
+}
+
+
+static int op_rpmb_program_key(int nargs, char *argv[])
+{
+	int ret;
+	int  dev_fd = -1, key_fd = -1;
+	uint16_t req = RPMB_PROGRAM_KEY;
+	struct rpmb_frame frame_in, frame_out;
+
+	ret = -EINVAL;
+	if (nargs != 2)
+		return ret;
+
+	dev_fd = open_dev_file(argv[0]);
+	if (dev_fd < 0)
+		goto out;
+	argv++;
+
+	key_fd = open_rd_file(argv[0], "key file");
+	if (key_fd < 0)
+		goto out;
+	argv++;
+
+	memset(&frame_in, 0, sizeof(frame_in));
+	frame_in.req_resp = htobe16(req);
+
+	read_file(key_fd, frame_in.key_mac, RPMB_KEY_SIZE);
+
+	ret = rpmb_ioctl(dev_fd, req, &frame_in, 1, &frame_out, 1);
+	if (ret)
+		goto out;
+
+	ret = be16toh(frame_out.result);
+	if (ret)
+		rpmb_err("RPMB operation %s failed, %s[0x%04x]\n",
+			 rpmb_op_str(req), rpmb_result_str(ret), ret);
+
+out:
+	close_fd(dev_fd);
+	close_fd(key_fd);
+
+	return ret;
+}
+
+static int rpmb_get_write_counter(int dev_fd, unsigned int *cnt,
+				  const unsigned char *key)
+{
+	int ret;
+	uint16_t req = RPMB_GET_WRITE_COUNTER;
+	unsigned char mac[RPMB_MAC_SIZE];
+	struct rpmb_frame frame_in;
+	struct rpmb_frame frame_out;
+
+	memset(&frame_in, 0, sizeof(frame_in));
+	frame_in.req_resp = htobe16(req);
+	RAND_bytes(frame_in.nonce, RPMB_NONCE_SIZE);
+
+	ret = rpmb_ioctl(dev_fd, req, &frame_in, 1, &frame_out, 1);
+	if (ret != 0)
+		return ret;
+
+	ret = be16toh(frame_out.result);
+	if (ret)
+		goto out;
+
+	if (key) {
+		rpmb_calc_hmac_sha256(&frame_out, 1,
+				      key, RPMB_KEY_SIZE,
+				      mac, RPMB_MAC_SIZE);
+		if (memcmp(mac, frame_out.key_mac, RPMB_MAC_SIZE)) {
+			rpmb_err("RPMB hmac mismatch\n");
+			dump_hex_buffer("Result MAC: ",
+					frame_out.key_mac, RPMB_MAC_SIZE);
+			dump_hex_buffer("Expected MAC: ", mac, RPMB_MAC_SIZE);
+			ret = RPMB_ERR_AUTH;
+			goto out;
+		}
+	}
+
+	*cnt = be32toh(frame_out.write_counter);
+
+out:
+	if (ret) {
+		rpmb_err("RPMB operation %s failed, %s[0x%04x]\n",
+			 rpmb_op_str(req), rpmb_result_str(ret), ret);
+	}
+	return ret;
+}
+
+static int op_rpmb_get_write_counter(int nargs, char **argv)
+{
+	int ret;
+	int dev_fd = -1, key_fd = -1;
+	bool has_key;
+	unsigned char key[RPMB_KEY_SIZE];
+	unsigned int cnt;
+
+	if (nargs == 2)
+		has_key = true;
+	else if (nargs == 1)
+		has_key = false;
+	else
+		return -EINVAL;
+
+	ret = -EINVAL;
+	dev_fd = open_dev_file(argv[0]);
+	if (dev_fd < 0)
+		return ret;
+	argv++;
+
+	if (has_key) {
+		key_fd = open_rd_file(argv[0], "key file");
+		if (key_fd < 0)
+			goto out;
+		argv++;
+
+		ret = read_file(key_fd, key, RPMB_KEY_SIZE);
+		if (ret < 0)
+			goto out;
+
+		ret = rpmb_get_write_counter(dev_fd, &cnt, key);
+	} else {
+		ret = rpmb_get_write_counter(dev_fd, &cnt, NULL);
+	}
+
+	if (!ret)
+		printf("Counter value: 0x%08x\n", cnt);
+
+out:
+	close_fd(dev_fd);
+	close_fd(key_fd);
+	return ret;
+}
+
+static int op_rpmb_read_blocks(int nargs, char **argv)
+{
+	int i, ret;
+	int dev_fd = -1, data_fd = -1, key_fd = -1;
+	uint16_t req = RPMB_READ_DATA;
+	uint16_t addr, blocks_cnt;
+	unsigned char key[RPMB_KEY_SIZE];
+	unsigned char mac[RPMB_MAC_SIZE];
+	unsigned long numarg;
+	bool has_key;
+	struct rpmb_frame frame_in;
+	struct rpmb_frame *frames_out = NULL;
+	struct rpmb_frame *frame_out;
+
+	ret = -EINVAL;
+	if (nargs == 4)
+		has_key = false;
+	else if (nargs == 5)
+		has_key = true;
+	else
+		return ret;
+
+	dev_fd = open_dev_file(argv[0]);
+	if (dev_fd < 0)
+		goto out;
+	argv++;
+
+	errno = 0;
+	numarg = strtoul(argv[0], NULL, 0);
+	if (errno || numarg > USHRT_MAX) {
+		rpmb_err("wrong block address\n");
+		goto out;
+	}
+	addr = (uint16_t)numarg;
+	argv++;
+
+	errno = 0;
+	numarg = strtoul(argv[0], NULL, 0);
+	if (errno || numarg > USHRT_MAX) {
+		rpmb_err("wrong blocks count\n");
+		goto out;
+	}
+	blocks_cnt = (uint16_t)numarg;
+	argv++;
+
+	if (blocks_cnt == 0) {
+		rpmb_err("wrong blocks count\n");
+		goto out;
+	}
+
+	data_fd = open_wr_file(argv[0], "output data");
+	if (data_fd < 0)
+		goto out;
+	argv++;
+
+	if (has_key) {
+		key_fd = open_rd_file(argv[0], "key file");
+		if (key_fd < 0)
+			goto out;
+		argv++;
+
+		ret = read_file(key_fd, key, RPMB_KEY_SIZE);
+		if (ret < 0)
+			goto out;
+	}
+
+	ret = 0;
+	memset(&frame_in, 0, sizeof(frame_in));
+	frame_in.req_resp = htobe16(req);
+	frame_in.addr = htobe16(addr);
+	frame_in.block_count = htobe16(blocks_cnt);
+	if (has_key)
+		RAND_bytes(frame_in.nonce, RPMB_NONCE_SIZE);
+
+	frames_out = calloc(blocks_cnt, sizeof(struct rpmb_frame));
+	if (!frames_out) {
+		rpmb_err("Cannot allocate %d RPMB frames\n", blocks_cnt);
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = rpmb_ioctl(dev_fd, req, &frame_in, 1, frames_out, blocks_cnt);
+	if (ret)
+		goto out;
+
+	frame_out = &frames_out[blocks_cnt - 1];
+	ret = be16toh(frame_out->result);
+	if (ret) {
+		rpmb_err("RPMB operation %s failed, %s[0x%04x]\n",
+			 rpmb_op_str(req), rpmb_result_str(ret), ret);
+		goto out;
+	}
+
+	if (has_key) {
+		rpmb_calc_hmac_sha256(frames_out, blocks_cnt,
+				      key, RPMB_KEY_SIZE,
+				      mac, RPMB_MAC_SIZE);
+		if (memcmp(mac, frame_out->key_mac, RPMB_MAC_SIZE)) {
+			rpmb_err("RPMB hmac mismatch\n");
+			dump_hex_buffer("Result MAC: ",
+					frame_out->key_mac, RPMB_MAC_SIZE);
+			dump_hex_buffer("Expected MAC: ", mac, RPMB_MAC_SIZE);
+			ret = RPMB_ERR_AUTH;
+			goto out;
+		}
+	}
+
+	for (i = 0; i < blocks_cnt; i++) {
+		ret = write_file(data_fd,
+				frames_out[i].data, sizeof(frames_out[i].data));
+		if (ret < 0)
+			goto out;
+	}
+
+out:
+	free(frames_out);
+	close_fd(dev_fd);
+	close_fd(data_fd);
+	close_fd(key_fd);
+
+	return ret;
+}
+
+static int op_rpmb_write_blocks(int nargs, char **argv)
+{
+	int ret;
+	int dev_fd = -1, key_fd = -1, data_fd = -1;
+	int i;
+	uint16_t req = RPMB_WRITE_DATA;
+	unsigned char key[RPMB_KEY_SIZE];
+	unsigned char mac[RPMB_MAC_SIZE];
+	unsigned long numarg;
+	uint16_t addr, blocks_cnt;
+	unsigned int cnt;
+	struct rpmb_frame *frames_in;
+	struct rpmb_frame frame_out;
+
+	ret = -EINVAL;
+	if (nargs != 5)
+		goto out;
+
+	dev_fd = open_dev_file(argv[0]);
+	if (dev_fd < 0)
+		goto out;
+	argv++;
+
+	errno = 0;
+	numarg = strtoul(argv[0], NULL, 0);
+	if (errno || numarg > USHRT_MAX) {
+		rpmb_err("wrong block address %s\n", argv[0]);
+		goto out;
+	}
+	addr = (uint16_t)numarg;
+	argv++;
+
+	errno = 0;
+	numarg = strtoul(argv[0], NULL, 0);
+	if (errno || numarg > USHRT_MAX) {
+		rpmb_err("wrong blocks count\n");
+		goto out;
+	}
+	blocks_cnt = (uint16_t)numarg;
+	argv++;
+
+	if (blocks_cnt == 0) {
+		rpmb_err("wrong blocks count\n");
+		goto out;
+	}
+
+	data_fd = open_rd_file(argv[0], "data file");
+	if (data_fd < 0)
+		goto out;
+	argv++;
+
+	key_fd = open_rd_file(argv[0], "key file");
+	if (key_fd < 0)
+		goto out;
+	argv++;
+
+	ret = read_file(key_fd, key, RPMB_KEY_SIZE);
+	if (ret < 0)
+		goto out;
+
+	frames_in = calloc(blocks_cnt, sizeof(struct rpmb_frame));
+	if (!frames_in) {
+		rpmb_err("can't allocate memory for RPMB outer frames\n");
+		exit(EXIT_FAILURE);
+	}
+
+	ret = rpmb_get_write_counter(dev_fd, &cnt, key);
+	if (ret)
+		goto out;
+
+	for (i = 0; i < blocks_cnt; i++) {
+		frames_in[i].req_resp      = htobe16(req);
+		frames_in[i].block_count   = htobe16(blocks_cnt);
+		frames_in[i].addr          = htobe16(addr);
+		frames_in[i].write_counter = htobe32(cnt);
+	}
+
+	for (i = 0; i < blocks_cnt; i++) {
+		ret = read_file(data_fd, frames_in[i].data,
+				sizeof(frames_in[0].data));
+		if (ret < 0)
+			goto out;
+	}
+
+	rpmb_calc_hmac_sha256(frames_in, blocks_cnt,
+			      key, RPMB_KEY_SIZE,
+			      mac, RPMB_MAC_SIZE);
+	memcpy(frames_in[blocks_cnt - 1].key_mac, mac, RPMB_MAC_SIZE);
+	ret = rpmb_ioctl(dev_fd, req, frames_in, blocks_cnt, &frame_out, 1);
+	if (ret != 0)
+		goto out;
+
+	ret = be16toh(frame_out.result);
+	if (ret) {
+		rpmb_err("RPMB operation %s failed, %s[0x%04x]\n",
+			 rpmb_op_str(req), rpmb_result_str(ret), ret);
+	}
+out:
+	close_fd(dev_fd);
+	close_fd(data_fd);
+	close_fd(key_fd);
+	return ret;
+}
+
+typedef int (*rpmb_op)(int argc, char *argv[]);
+
+struct rpmb_cmd {
+	const char *op_name;
+	rpmb_op     op;
+	const char  *usage; /* usage title */
+	const char  *help;  /* help */
+};
+
+
+static const struct rpmb_cmd cmds[] = {
+	{
+	 "program-key",
+	 op_rpmb_program_key,
+	 "<RPMB_DEVICE> <KEY_FILE>",
+	 "    Program authentication key of 32 bytes length from the KEY_FILE\n"
+	 "    when KEY_FILE is -, read standard input.\n"
+	 "    NOTE: This is a one-time programmable irreversible change.\n",
+	},
+	{
+	 "write-counter",
+	 op_rpmb_get_write_counter,
+	 "<RPMB_DEVICE> [KEY_FILE]",
+	 "    Rertrive write counter value from the <RPMB_DEVICE> to stdout.\n"
+	 "    When KEY_FILE is present data is verified via HMAC\n"
+	 "    when KEY_FILE is -, read standard input.\n"
+	},
+	{
+	  "write-blocks",
+	  op_rpmb_write_blocks,
+	  "<RPMB_DEVICE> <address> <block_count> <DATA_FILE> <KEY_FILE>",
+	  "    <block count> of 256 bytes will be written from the DATA_FILE\n"
+	  "    to the <RPMB_DEVICE> at block offset <address>.\n"
+	  "    When DATA_FILE is -, read from standard input.\n",
+	},
+	{
+	  "read-blocks",
+	  op_rpmb_read_blocks,
+	  "<RPMB_DEVICE> <address> <blocks count> <OUTPUT_FILE> [KEY_FILE]",
+	  "    <block count> of 256 bytes will be read from <RPMB_DEVICE>\n"
+	  "    to the OUTPUT_FILE\n"
+	  "    When KEY_FILE is present data is verified via HMAC\n"
+	  "    When OUTPUT/KEY_FILE is -, read from standard input.\n"
+	  "    When OUTPUT_FILE is -, write to standard output\n",
+	},
+
+	{ NULL, NULL, NULL, NULL }
+};
+
+static void help(const char *prog, const struct rpmb_cmd *cmd)
+{
+	printf("%s %s %s\n", prog, cmd->op_name, cmd->usage);
+	printf("%s\n", cmd->help);
+}
+
+static void usage(const char *prog)
+{
+	int i;
+
+	printf("\n");
+	printf("Usage: %s [-v] <command> <args>\n\n", prog);
+	for (i = 0; cmds[i].op_name; i++)
+		printf("       %s %s %s\n",
+			prog, cmds[i].op_name, cmds[i].usage);
+
+	printf("\n");
+	printf("      %s -v: runs in verbose mode\n", prog);
+	printf("      %s help : shows this help\n", prog);
+	printf("      %s help <command>: shows detailed help\n", prog);
+}
+
+static bool call_for_help(const char *arg)
+{
+	return !strcmp(arg, "help") ||
+	       !strcmp(arg, "-h")   ||
+	       !strcmp(arg, "--help");
+}
+
+static bool parse_verbose(const char *arg)
+{
+	return !strcmp(arg, "-v") ||
+	       !strcmp(arg, "--verbose");
+}
+
+static const
+struct rpmb_cmd *parse_args(const char *prog, int *_argc, char **_argv[])
+{
+
+	int i;
+	int argc = *_argc;
+	char **argv =  *_argv;
+	const struct rpmb_cmd *cmd = NULL;
+	bool need_help = false;
+
+	argc--; argv++;
+
+	if (argc == 0)
+		goto out;
+
+	if (call_for_help(argv[0])) {
+		argc--; argv++;
+		if (argc == 0)
+			goto out;
+		else
+			need_help = true;
+	}
+
+	if (parse_verbose(argv[0])) {
+		argc--; argv++;
+		if (argc == 0)
+			goto out;
+		else
+			verbose = true;
+	}
+
+	for (i = 0; cmds[i].op_name; i++) {
+		if (!strncmp(argv[0], cmds[i].op_name,
+			     strlen(cmds[i].op_name))) {
+			cmd = &cmds[i];
+			argc--; argv++;
+			break;
+		}
+	}
+
+	if (!cmd)
+		goto out;
+
+	if (need_help || (argc > 0 && call_for_help(argv[0]))) {
+		help(prog, cmd);
+		argc--; argv++;
+		return NULL;
+	}
+
+out:
+	*_argc = argc;
+	*_argv = argv;
+
+	if (!cmd)
+		usage(prog);
+
+	return cmd;
+}
+
+int main(int argc, char *argv[])
+{
+	const char *prog = basename(argv[0]);
+	const struct rpmb_cmd *cmd;
+	int ret;
+
+	cmd = parse_args(prog, &argc, &argv);
+	if (!cmd)
+		exit(EXIT_SUCCESS);
+
+	ret = cmd->op(argc, argv);
+	if (ret == -EINVAL)
+		help(prog, cmd);
+
+	if (ret)
+		exit(EXIT_FAILURE);
+
+	exit(EXIT_SUCCESS);
+}
-- 
2.5.5

[PATCH v4 7/8] mmc: block: register RPMB partition with the RPMB subsystem

[Tomas Winkler]

Register eMMC RPMB partition with the RPMB subsystem and provide
implementation for the RPMB access operations abstracting
actual multi step process.

V2: resend
V3: commit message fix
V4: Kconfig: use select RPMB to ensure valid configuration
    Switch back to main area after RPMB access

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
---
 drivers/mmc/card/Kconfig |   1 +
 drivers/mmc/card/block.c | 292 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 293 insertions(+)

diff --git a/drivers/mmc/card/Kconfig b/drivers/mmc/card/Kconfig
index 5562308699bc..537d0bc82781 100644
--- a/drivers/mmc/card/Kconfig
+++ b/drivers/mmc/card/Kconfig
@@ -7,6 +7,7 @@ comment "MMC/SD/SDIO Card Drivers"
 config MMC_BLOCK
 	tristate "MMC block device driver"
 	depends on BLOCK
+	select RPMB
 	default y
 	help
 	  Say Y here to enable the MMC block device driver support.
diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
index e62fde3ac431..274a21fe8be5 100644
--- a/drivers/mmc/card/block.c
+++ b/drivers/mmc/card/block.c
@@ -42,6 +42,7 @@
 #include <linux/mmc/host.h>
 #include <linux/mmc/mmc.h>
 #include <linux/mmc/sd.h>
+#include <linux/rpmb.h>
 
 #include <asm/uaccess.h>
 
@@ -110,6 +111,7 @@ struct mmc_blk_data {
 #define MMC_BLK_WRITE		BIT(1)
 #define MMC_BLK_DISCARD		BIT(2)
 #define MMC_BLK_SECDISCARD	BIT(3)
+#define MMC_BLK_RPMB		BIT(4)
 
 	/*
 	 * Only set in main mmc_blk_data associated
@@ -1162,6 +1164,292 @@ int mmc_access_rpmb(struct mmc_queue *mq)
 	return false;
 }
 
+static int mmc_rpmb_send_cmd(struct mmc_card *card,
+			     u16 rpmb_type, int data_type,
+			     u8 *buf, u16 blks)
+{
+	struct mmc_command sbc = {
+		.opcode = MMC_SET_BLOCK_COUNT,
+		.flags  = MMC_RSP_SPI_R1 | MMC_RSP_R1 | MMC_CMD_AC,
+	};
+
+	struct mmc_command cmd = {
+		.flags = MMC_RSP_SPI_R1 | MMC_RSP_R1 | MMC_CMD_ADTC,
+	};
+
+	struct mmc_data data = {
+		.blksz = 512,
+	};
+	struct mmc_request mrq = {
+		.sbc    = &sbc,
+		.cmd    = &cmd,
+		.data   = &data,
+		.stop   = NULL,
+	};
+	struct scatterlist sg;
+	bool do_rel_wr;
+
+	/*  set CMD23 */
+	sbc.arg = blks & 0x0000FFFF;
+	do_rel_wr = (rpmb_type == RPMB_WRITE_DATA ||
+		     rpmb_type == RPMB_PROGRAM_KEY);
+
+	if (do_rel_wr)
+		sbc.arg |= MMC_CMD23_ARG_REL_WR;
+
+	/*  set CMD25/18 */
+	cmd.opcode = (data_type == MMC_DATA_WRITE) ?
+		MMC_WRITE_MULTIPLE_BLOCK : MMC_READ_MULTIPLE_BLOCK;
+
+	sg_init_one(&sg, buf, 512 * blks);
+
+	data.blocks = blks;
+	data.sg     = &sg;
+	data.sg_len = 1;
+	data.flags  = data_type;
+	mmc_set_data_timeout(&data, card);
+
+	mmc_wait_for_req(card->host, &mrq);
+
+	if (cmd.error) {
+		pr_err("%s: %s cmd error (%d)\n",
+			mmc_hostname(card->host), __func__, cmd.error);
+		return cmd.error;
+	}
+	if (data.error) {
+		pr_err("%s: %s data error (%d)\n",
+			mmc_hostname(card->host), __func__, data.error);
+		return data.error;
+	}
+	return 0;
+}
+
+static int mmc_blk_rpmb_sequence(struct mmc_card *card,
+			    struct rpmb_data *rpmbd)
+{
+	struct rpmb_frame *in_frames, *out_frames;
+	u8 *in_buf, *out_buf;
+	u16 blks;
+	u16 type;
+	int err;
+
+	in_frames = rpmbd->in_frames;
+	out_frames = rpmbd->out_frames;
+	in_buf = (u8 *)in_frames;
+	out_buf = (u8 *)out_frames;
+
+	type = rpmbd->req_type;
+	blks = be16_to_cpu(in_frames[0].block_count);
+
+	switch (type) {
+	case RPMB_PROGRAM_KEY:
+		blks = 1;
+		/* fall through */
+	case RPMB_WRITE_DATA:
+		/* STEP 1: send request to RPMB partition */
+		err = mmc_rpmb_send_cmd(card, type, MMC_DATA_WRITE,
+					in_buf, blks);
+		if (err) {
+			pr_err("%s: mmc_rpmb_send_cmd failed(%d)\n",
+				mmc_hostname(card->host), err);
+			goto out;
+		}
+
+		/* STEP 2: check write result (reuse out_frames) */
+		memset(out_frames, 0, 512);
+		out_frames[0].req_resp = cpu_to_be16(RPMB_RESULT_READ);
+		err = mmc_rpmb_send_cmd(card,
+			RPMB_RESULT_READ, MMC_DATA_WRITE, out_buf, 1);
+		if (err) {
+			pr_err("%s: mmc_rpmb_send_cmd failed(%d)\n",
+				mmc_hostname(card->host), err);
+			goto out;
+		}
+
+		/* STEP 3: get response from RPMB partition */
+		err = mmc_rpmb_send_cmd(card,
+			RPMB_READ_DATA, MMC_DATA_READ, out_buf, 1);
+
+		if (err) {
+			pr_err("%s: mmc_rpmb_send_cmd failed(%d)\n",
+				mmc_hostname(card->host), err);
+			goto out;
+		}
+		break;
+
+	case RPMB_GET_WRITE_COUNTER:
+		blks = 1;
+		/* fall through */
+	case RPMB_READ_DATA:
+		/* STEP 1: send request to RPMB partition */
+		err = mmc_rpmb_send_cmd(card, type, MMC_DATA_WRITE, in_buf, 1);
+		if (err) {
+			pr_err("%s: mmc_rpmb_send_cmd failed(%d)\n",
+				mmc_hostname(card->host), err);
+			goto out;
+		}
+
+		/* STEP 3: get response from RPMB partition */
+		err = mmc_rpmb_send_cmd(card, type, MMC_DATA_READ,
+					out_buf, blks);
+		if (err) {
+			pr_err("%s: mmc_rpmb_send_cmd failed(%d)\n",
+				mmc_hostname(card->host), err);
+			goto out;
+		}
+		break;
+
+	default:
+		err = -EINVAL;
+		goto out;
+	}
+out:
+	return err;
+}
+
+static int mmc_blk_rpmb_req_process(struct mmc_blk_data *md,
+				    struct rpmb_data *rpmbd)
+{
+	struct mmc_card *card;
+	int ret;
+
+	if (WARN_ON(!md || !rpmbd))
+		return -EINVAL;
+
+	if (!(md->flags & MMC_BLK_CMD23) ||
+	     (md->part_type != EXT_CSD_PART_CONFIG_ACC_RPMB))
+		return -EOPNOTSUPP;
+
+	card = md->queue.card;
+	if (!card || !mmc_card_mmc(card))
+		return -ENODEV;
+
+
+	pr_debug("%s rpmb request type = 0x%1x\n",
+		md->disk->disk_name, rpmbd->req_type);
+
+	mmc_get_card(card);
+
+	/* switch to RPMB partition */
+	ret = mmc_blk_part_switch(card, md);
+	if (ret) {
+		pr_err("%s: Invalid RPMB partition switch (%d)!\n",
+			mmc_hostname(card->host), ret);
+		/*
+		 * In case partition is not in user data area, make
+		 * a force partition switch.
+		 * we need reset eMMC card at here
+		 */
+		ret = mmc_blk_reset(md, card->host, MMC_BLK_RPMB);
+		if (!ret)
+			mmc_blk_reset_success(md, MMC_BLK_RPMB);
+		else
+			pr_err("%s: eMMC card reset failed (%d)\n",
+				mmc_hostname(card->host), ret);
+		goto out;
+	}
+
+	ret = mmc_blk_rpmb_sequence(card, rpmbd);
+	if (ret)
+		pr_err("%s: failed (%d) to handle RPMB request type %d!\n",
+			mmc_hostname(card->host), ret, rpmbd->req_type);
+
+	/* Always switch back to main area after RPMB access */
+	mmc_blk_part_switch(card, dev_get_drvdata(&card->dev));
+out:
+	mmc_put_card(card);
+	return ret;
+}
+
+static int mmc_blk_rpmb_send_req(struct device *dev, struct rpmb_data *req)
+{
+	struct mmc_blk_data *md = mmc_blk_get(dev_to_disk(dev));
+	int ret;
+
+	if (!md)
+		return -ENODEV;
+
+	ret = mmc_blk_rpmb_req_process(md, req);
+
+	mmc_blk_put(md);
+
+	return ret;
+}
+
+static struct rpmb_ops mmc_rpmb_dev_ops = {
+	.send_rpmb_req = mmc_blk_rpmb_send_req,
+	.type = RPMB_TYPE_EMMC,
+};
+
+static struct mmc_blk_data *mmc_blk_rpmb_part_get(struct mmc_blk_data *md)
+{
+	struct mmc_blk_data *part_md;
+
+	if (!md)
+		return NULL;
+
+	list_for_each_entry(part_md, &md->part, part) {
+		if (part_md->area_type == MMC_BLK_DATA_AREA_RPMB)
+			return part_md;
+	}
+
+	return NULL;
+}
+
+static void mmc_blk_rpmb_unset_dev_id(struct rpmb_ops *ops)
+{
+	kfree(ops->dev_id);
+	ops->dev_id = NULL;
+}
+
+static int mmc_blk_rpmb_set_dev_id(struct rpmb_ops *ops, struct mmc_card *card)
+{
+	char *id;
+
+	id = kmalloc(sizeof(card->raw_cid), GFP_KERNEL);
+	if (!id)
+		return -ENOMEM;
+
+	memcpy(id, card->raw_cid, sizeof(card->raw_cid));
+	ops->dev_id = id;
+	ops->dev_id_len = sizeof(card->raw_cid);
+
+	return 0;
+}
+
+static void mmc_blk_rpmb_add(struct mmc_card *card)
+{
+	struct mmc_blk_data *md = dev_get_drvdata(&card->dev);
+	struct mmc_blk_data *part_md = mmc_blk_rpmb_part_get(md);
+	struct rpmb_dev *rdev;
+
+	if (!part_md)
+		return;
+
+	mmc_blk_rpmb_set_dev_id(&mmc_rpmb_dev_ops, card);
+
+	/* RPMB blocks are written in half sectors hence '* 2' */
+	mmc_rpmb_dev_ops.reliable_wr_cnt = card->ext_csd.rel_sectors * 2;
+
+	rdev = rpmb_dev_register(disk_to_dev(part_md->disk),
+				  &mmc_rpmb_dev_ops);
+	if (IS_ERR(rdev)) {
+		pr_warn("%s: cannot register to rpmb %ld\n",
+			part_md->disk->disk_name, PTR_ERR(rdev));
+	}
+}
+
+static void mmc_blk_rpmb_remove(struct mmc_card *card)
+{
+	struct mmc_blk_data *md = dev_get_drvdata(&card->dev);
+	struct mmc_blk_data *part_md = mmc_blk_rpmb_part_get(md);
+
+	if (part_md)
+		rpmb_dev_unregister(disk_to_dev(part_md->disk));
+
+	mmc_blk_rpmb_unset_dev_id(&mmc_rpmb_dev_ops);
+}
+
 static int mmc_blk_issue_discard_rq(struct mmc_queue *mq, struct request *req)
 {
 	struct mmc_blk_data *md = mq->data;
@@ -2613,6 +2901,8 @@ static int mmc_blk_probe(struct mmc_card *card)
 			goto out;
 	}
 
+	mmc_blk_rpmb_add(card);
+
 	pm_runtime_set_autosuspend_delay(&card->dev, 3000);
 	pm_runtime_use_autosuspend(&card->dev);
 
@@ -2637,6 +2927,7 @@ static void mmc_blk_remove(struct mmc_card *card)
 {
 	struct mmc_blk_data *md = dev_get_drvdata(&card->dev);
 
+	mmc_blk_rpmb_remove(card);
 	mmc_blk_remove_parts(card, md);
 	pm_runtime_get_sync(&card->dev);
 	mmc_claim_host(card->host);
@@ -2665,6 +2956,7 @@ static int _mmc_blk_suspend(struct mmc_card *card)
 
 static void mmc_blk_shutdown(struct mmc_card *card)
 {
+	mmc_blk_rpmb_remove(card);
 	_mmc_blk_suspend(card);
 }
 
-- 
2.5.5

[PATCH v4 8/8] scsi: ufs: connect to RPMB subsystem

[Tomas Winkler]

Register UFS RPMB LUN with the RPMB subsystem and provide
implementation for the RPMB access operations. RPMB partition is
accessed via a sequence of security protocol in and security protocol
out commands with UFS specific parameters. This multi step process is
abstracted into 4 basic RPMB commands.

V2: resend
V3: resend
V4: Kconfig: use select RPMB to ensure valid configuration

Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
 drivers/scsi/ufs/Kconfig  |   1 +
 drivers/scsi/ufs/ufshcd.c | 219 ++++++++++++++++++++++++++++++++++++++++++++++
 drivers/scsi/ufs/ufshcd.h |   2 +
 3 files changed, 222 insertions(+)

diff --git a/drivers/scsi/ufs/Kconfig b/drivers/scsi/ufs/Kconfig
index 097894a1fab5..e6ff8ecff695 100644
--- a/drivers/scsi/ufs/Kconfig
+++ b/drivers/scsi/ufs/Kconfig
@@ -38,6 +38,7 @@ config SCSI_UFSHCD
 	select PM_DEVFREQ
 	select DEVFREQ_GOV_SIMPLE_ONDEMAND
 	select NLS
+	select RPMB
 	---help---
 	This selects the support for UFS devices in Linux, say Y and make
 	  sure that you know the name of your UFS host adapter (the card
diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index f8fa72c31a9d..c087e9713db3 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -37,10 +37,13 @@
  * license terms, and distributes only under these terms.
  */
 
+#include <asm/unaligned.h>
 #include <linux/async.h>
 #include <linux/devfreq.h>
 #include <linux/nls.h>
 #include <linux/of.h>
+#include <linux/rpmb.h>
+
 #include "ufshcd.h"
 #include "ufs_quirks.h"
 #include "unipro.h"
@@ -4744,6 +4747,214 @@ static void ufshcd_init_icc_levels(struct ufs_hba *hba)
 
 }
 
+#define SEC_PROTOCOL_UFS  0xEC
+#define   SEC_SPECIFIC_UFS_RPMB 0x0001
+
+#define SEC_PROTOCOL_CMD_SIZE 12
+#define SEC_PROTOCOL_RETRIES 3
+#define SEC_PROTOCOL_RETRIES_ON_RESET 10
+#define SEC_PROTOCOL_TIMEOUT msecs_to_jiffies(1000)
+
+static int
+ufshcd_rpmb_security_out(struct scsi_device *sdev,
+			 struct rpmb_frame *frames, u32 cnt)
+{
+	struct scsi_sense_hdr sshdr;
+	u32 trans_len = cnt * sizeof(struct rpmb_frame);
+	int reset_retries = SEC_PROTOCOL_RETRIES_ON_RESET;
+	int ret;
+	u8 cmd[SEC_PROTOCOL_CMD_SIZE];
+
+retry:
+	memset(cmd, 0, SEC_PROTOCOL_CMD_SIZE);
+	cmd[0] = SECURITY_PROTOCOL_OUT;
+	cmd[1] = SEC_PROTOCOL_UFS;
+	put_unaligned_be16(SEC_SPECIFIC_UFS_RPMB, cmd + 2);
+	cmd[4] = 0;                              /* inc_512 bit 7 set to 0 */
+	put_unaligned_be32(trans_len, cmd + 6);  /* transfer length */
+
+	ret = scsi_execute_req_flags(sdev, cmd, DMA_TO_DEVICE,
+				     frames, trans_len, &sshdr,
+				     SEC_PROTOCOL_TIMEOUT, SEC_PROTOCOL_RETRIES,
+				     NULL, 0);
+	if (ret && scsi_sense_valid(&sshdr) &&
+	    sshdr.sense_key == UNIT_ATTENTION &&
+	    sshdr.asc == 0x29 && sshdr.ascq == 0x00)
+		/* Device reset might occur several times,
+		 * give it one more chance
+		 */
+		if (--reset_retries > 0)
+			goto retry;
+
+	if (ret)
+		pr_err("%s: failed with err %0x\n", __func__, ret);
+
+	if (driver_byte(ret) & DRIVER_SENSE)
+		scsi_print_sense_hdr(sdev, "rpmb: security out", &sshdr);
+
+	return ret;
+}
+
+static int
+ufshcd_rpmb_security_in(struct scsi_device *sdev,
+			struct rpmb_frame *frames, u32 cnt)
+{
+	struct scsi_sense_hdr sshdr;
+	u32 alloc_len = cnt * sizeof(struct rpmb_frame);
+	int reset_retries = SEC_PROTOCOL_RETRIES_ON_RESET;
+	int ret;
+	u8 cmd[SEC_PROTOCOL_CMD_SIZE];
+
+retry:
+	memset(cmd, 0, SEC_PROTOCOL_CMD_SIZE);
+	cmd[0] = SECURITY_PROTOCOL_IN;
+	cmd[1] = SEC_PROTOCOL_UFS;
+	put_unaligned_be16(SEC_SPECIFIC_UFS_RPMB, cmd + 2);
+	cmd[4] = 0;                             /* inc_512 bit 7 set to 0 */
+	put_unaligned_be32(alloc_len, cmd + 6); /* allocation length */
+
+	ret = scsi_execute_req_flags(sdev, cmd, DMA_FROM_DEVICE,
+				     frames, alloc_len, &sshdr,
+				     SEC_PROTOCOL_TIMEOUT, SEC_PROTOCOL_RETRIES,
+				     NULL, 0);
+	if (ret && scsi_sense_valid(&sshdr) &&
+	    sshdr.sense_key == UNIT_ATTENTION &&
+	    sshdr.asc == 0x29 && sshdr.ascq == 0x00)
+		/* Device reset might occur several times,
+		 * give it one more chance
+		 */
+		if (--reset_retries > 0)
+			goto retry;
+
+	if (ret)
+		pr_err("%s: failed with err %0x\n", __func__, ret);
+
+	if (driver_byte(ret) & DRIVER_SENSE)
+		scsi_print_sense_hdr(sdev, "rpmb: security in", &sshdr);
+
+	return ret;
+}
+
+
+static int ufshcd_rpmb_send_req(struct device *dev, struct rpmb_data *rpmbd)
+{
+	unsigned long flags;
+	struct ufs_hba *hba = dev_get_drvdata(dev);
+	struct scsi_device *sdev;
+	struct rpmb_frame *in_frames, *out_frames;
+	u16 blks;
+	u16 type;
+	int ret;
+
+	in_frames = rpmbd->in_frames;
+	out_frames = rpmbd->out_frames;
+
+	type = rpmbd->req_type;
+	blks = be16_to_cpu(in_frames[0].block_count);
+
+	dev_dbg(hba->dev, "RPMB : type = %d, blocks = %d\n", type, blks);
+
+	spin_lock_irqsave(hba->host->host_lock, flags);
+	sdev = hba->sdev_ufs_rpmb;
+	if (sdev) {
+		ret = scsi_device_get(sdev);
+		if (!ret && !scsi_device_online(sdev)) {
+			ret = -ENODEV;
+			scsi_device_put(sdev);
+		}
+	} else {
+		ret = -ENODEV;
+	}
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+	if (ret)
+		return ret;
+
+	switch (type) {
+	case RPMB_PROGRAM_KEY:
+		blks = 1;
+		/* fall through */
+	case RPMB_WRITE_DATA:
+		/* STEP 1: send request to RPMB partition */
+		ret = ufshcd_rpmb_security_out(sdev, in_frames, blks);
+		if (ret)
+			break;
+
+		/* STEP 2: check write result (reuse out_frames) */
+		memset(out_frames, 0, 512);
+		out_frames[0].req_resp = cpu_to_be16(RPMB_RESULT_READ);
+		ret = ufshcd_rpmb_security_out(sdev, out_frames, 1);
+		if (ret)
+			break;
+
+		/* STEP 3: get response from RPMB partition */
+		ret = ufshcd_rpmb_security_in(sdev, out_frames, 1);
+		if (ret)
+			break;
+
+		break;
+	case RPMB_GET_WRITE_COUNTER:
+		blks = 1;
+		/* fall through */
+	case RPMB_READ_DATA:
+		/* STEP 1: send request to RPMB partition */
+		ret = ufshcd_rpmb_security_out(sdev, in_frames, 1);
+		if (ret)
+			break;
+		/* STEP 2: get response from RPMB partition */
+		ret = ufshcd_rpmb_security_in(sdev, out_frames, blks);
+		if (ret)
+			break;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	scsi_device_put(sdev);
+	return ret;
+}
+
+static struct rpmb_ops ufshcd_rpmb_dev_ops = {
+	.send_rpmb_req = ufshcd_rpmb_send_req,
+	.type = RPMB_TYPE_UFS,
+};
+
+static inline void ufshcd_rpmb_add(struct ufs_hba *hba)
+{
+	struct rpmb_dev *rdev;
+
+	scsi_device_get(hba->sdev_ufs_rpmb);
+	rdev = rpmb_dev_register(hba->dev, &ufshcd_rpmb_dev_ops);
+	if (IS_ERR(rdev)) {
+		dev_warn(hba->dev, "%s: cannot register to rpmb %ld\n",
+			 dev_name(hba->dev), PTR_ERR(rdev));
+		goto out_put_dev;
+	}
+
+	return;
+
+out_put_dev:
+	scsi_device_put(hba->sdev_ufs_rpmb);
+	hba->sdev_ufs_rpmb = NULL;
+}
+
+static inline void ufshcd_rpmb_remove(struct ufs_hba *hba)
+{
+	unsigned long flags;
+
+	if (!hba->sdev_ufs_rpmb)
+		return;
+
+	spin_lock_irqsave(hba->host->host_lock, flags);
+
+	rpmb_dev_unregister(hba->dev);
+	scsi_device_put(hba->sdev_ufs_rpmb);
+	hba->sdev_ufs_rpmb = NULL;
+
+	spin_unlock_irqrestore(hba->host->host_lock, flags);
+}
+
+
 /**
  * ufshcd_scsi_add_wlus - Adds required W-LUs
  * @hba: per-adapter instance
@@ -4799,7 +5010,11 @@ static int ufshcd_scsi_add_wlus(struct ufs_hba *hba)
 		ret = PTR_ERR(sdev_rpmb);
 		goto remove_sdev_boot;
 	}
+	hba->sdev_ufs_rpmb = sdev_rpmb;
+
+	ufshcd_rpmb_add(hba);
 	scsi_device_put(sdev_rpmb);
+
 	goto out;
 
 remove_sdev_boot:
@@ -6168,6 +6383,8 @@ int ufshcd_shutdown(struct ufs_hba *hba)
 			goto out;
 	}
 
+	ufshcd_rpmb_remove(hba);
+
 	ret = ufshcd_suspend(hba, UFS_SHUTDOWN_PM);
 out:
 	if (ret)
@@ -6184,6 +6401,8 @@ EXPORT_SYMBOL(ufshcd_shutdown);
  */
 void ufshcd_remove(struct ufs_hba *hba)
 {
+	ufshcd_rpmb_remove(hba);
+
 	scsi_remove_host(hba->host);
 	/* disable interrupts */
 	ufshcd_disable_intr(hba, hba->intr_mask);
diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h
index 4bb65669f052..92bfddf80ae8 100644
--- a/drivers/scsi/ufs/ufshcd.h
+++ b/drivers/scsi/ufs/ufshcd.h
@@ -346,6 +346,7 @@ struct ufs_init_prefetch {
  * @utmrdl_dma_addr: UTMRDL DMA address
  * @host: Scsi_Host instance of the driver
  * @dev: device handle
+ * @sdev_ufs_rpmb: reference to RPMB device W-LU
  * @lrb: local reference block
  * @lrb_in_use: lrb in use
  * @outstanding_tasks: Bits representing outstanding task requests
@@ -408,6 +409,7 @@ struct ufs_hba {
 	 * "UFS device" W-LU.
 	 */
 	struct scsi_device *sdev_ufs_device;
+	struct scsi_device *sdev_ufs_rpmb;
 
 	enum ufs_dev_pwr_mode curr_dev_pwr_mode;
 	enum uic_link_state uic_link_state;
-- 
2.5.5

Re: [PATCH v4 0/8] Replay Protected Memory Block (RPMB) subsystem

[Arve Hjønnevåg]

On Wed, Jun 1, 2016 at 2:41 PM, Tomas Winkler <tomas.winkler@intel.com> wrote:
> Few storage technology such is EMMC, UFS, and NVMe support RPMB
> hardware partition with common protocol and frame layout.
> The RPMB partition cannot be accessed via standard block layer, but
> by a set of specific commands: WRITE, READ, GET_WRITE_COUNTER, and
> PROGRAM_KEY.
>...

If the same protocol is used by all these standards, why not export it
directly (including the RESULT_READ command or not even knowing the
command types)? While I would prefer an rpmb specific interface over
the existing raw mmc command interface, all I need is an rpmb
operation that lets me send and receive buffers without interruption.
You can find our exiting user-space code here at
https://android.googlesource.com/platform/system/core/+/master/trusty/storage/proxy/rpmb.c.
If you use an interface more similar to this, I think your emmc and
ufs specific code would be simpler. Also, if you don't need the
in-kernel interface, the kernel would not need to know the details of
the rpmb protocol at all.

I have not tested your code, but it looks like we would have to modify
the storage proxy to interpret the data it currently passes through
and remove all RESULT_READ packets.

-- 
Arve Hjønnevåg

RE: [PATCH v4 0/8] Replay Protected Memory Block (RPMB) subsystem

[Winkler, Tomas]

> 
> On Wed, Jun 1, 2016 at 2:41 PM, Tomas Winkler <tomas.winkler@intel.com>
> wrote:
> > Few storage technology such is EMMC, UFS, and NVMe support RPMB
> >hardware partition with common protocol and frame layout.
> > The RPMB partition cannot be accessed via standard block layer, but
> >by a set of specific commands: WRITE, READ, GET_WRITE_COUNTER, and
> >PROGRAM_KEY.
> >...
> 
> If the same protocol is used by all these standards, why not export it directly
> (including the RESULT_READ command or not even knowing the command
> types)?
The protocol is the  same, but the wrapping of the packets is storage type specific so 
I believe that the best abstraction is on those  4 operations level. I'm not sure if the code would
be simpler if it would be exposed on a lower level.  
RESULT_READ  is  a command to be issued for getting preceeding write operation status, it's a kind of work around about the fact that you have to issue a read operation
in order to retrieve data in this case a  write operation result.  It can be successfully hidden and final result of the operation is delivered
to the user.

> While I would prefer an rpmb specific interface over the existing raw
> mmc command interface, all I need is an rpmb operation that lets me send
> and receive buffers without interruption.

I currently don't see an obstacle on doing the same with proposed interface, what is the issue are seeing.

> You can find our exiting user-space code here at
> https://android.googlesource.com/platform/system/core/+/master/trusty/s
> torage/proxy/rpmb.c.
> If you use an interface more similar to this, I think your emmc and ufs
> specific code would be simpler. Also, if you don't need the in-kernel
> interface, the kernel would not need to know the details of the rpmb
> protocol at all.

My major interest is the in-kernel protocol the user space API was more intended for debug, but I found it would be even more useful. 
The store type  access is very RPMB specific  for both MMC and UFS and needs to do special operations for RPMB so I don't see how this awareness can be removed or I'm not reading your proposal correctly.
Accessing RPMB is a multiple stepsoperation, the steps can be driven from the user space as done in EMMC ioctl but hidning would reduce the number of system calls and amount of data passed, 
in some sense the same as in the  new mmc  MMC_IOC_MULTI_CMD is trying ot achive.

> I have not tested your code, but it looks like we would have to modify the
> storage proxy to interpret the data it currently passes through and remove all
> RESULT_READ packets.
Your proxy driver just sends ioctls so this wouldn't be much difference and the rpmb code on the trusty w need rewrite just rpmb_send () function,
actually removing one set of buffer size. I will try to make that modification if it helps?

Thanks for interest and your review. 
Tomas 
 

Re: [v4,4/8] char: rpmb: provide user space interface

[Jérôme Forissier]

Hi Tomas,

On 06/01/2016 11:41 PM, Winkler, Tomas wrote:
> The user space API is achieved via single synchronous IOCTL.
> The request is submitted in in_frames_ptr pointer and received
> in out_frames_ptr.

I like this, because it's exactly what I wish existed when I wrote the
eMMC RPMB back-end for OP-TEE secure storage [1] ;-)

Same goes for patch 5/8 "char: rpmb: add RPMB simulation device": with
this I could just drop my custom simulation layer.

So, this goes in the right direction IMHO.

[1]
https://github.com/OP-TEE/optee_client/blob/master/tee-supplicant/src/rpmb.c

-- 
Jerome

Re: [PATCH v4 0/8] Replay Protected Memory Block (RPMB) subsystem

[Arve Hjønnevåg]

On Thu, Jun 2, 2016 at 6:17 AM, Winkler, Tomas <tomas.winkler@intel.com> wrote:
>>
>> On Wed, Jun 1, 2016 at 2:41 PM, Tomas Winkler <tomas.winkler@intel.com>
>> wrote:
>> > Few storage technology such is EMMC, UFS, and NVMe support RPMB
>> >hardware partition with common protocol and frame layout.
>> > The RPMB partition cannot be accessed via standard block layer, but
>> >by a set of specific commands: WRITE, READ, GET_WRITE_COUNTER, and
>> >PROGRAM_KEY.
>> >...
>>
>> If the same protocol is used by all these standards, why not export it directly
>> (including the RESULT_READ command or not even knowing the command
>> types)?
> The protocol is the  same, but the wrapping of the packets is storage type specific so
> I believe that the best abstraction is on those  4 operations level. I'm not sure if the code would
> be simpler if it would be exposed on a lower level.

I disagree. With the two storage types you support, the packets are
identical. The only difference is the low level commands you use to
send and receive them.

> RESULT_READ  is  a command to be issued for getting preceeding write operation status, it's a kind of work around about the fact that you have to issue a read operation
> in order to retrieve data in this case a  write operation result.  It can be successfully hidden and final result of the operation is delivered
> to the user.
>

Yes, it is possible to hide the result read command, but that does not
mean you should. The rpmb protocol is designed to let two endpoints
communicate in a way that lets them detect tampering. While the packet
you inject does not contain any protected data, you can still view it
as a form of tampering. If a future rpmb protocol version adds
features, you could loose the ability to inject packets.

>> While I would prefer an rpmb specific interface over the existing raw
>> mmc command interface, all I need is an rpmb operation that lets me send
>> and receive buffers without interruption.
>
> I currently don't see an obstacle on doing the same with proposed interface, what is the issue are seeing.
>

The main issue is that you are injecting commands, so code that
follows the mmc spec will not work.

>> You can find our exiting user-space code here at
>> https://android.googlesource.com/platform/system/core/+/master/trusty/s
>> torage/proxy/rpmb.c.
>> If you use an interface more similar to this, I think your emmc and ufs
>> specific code would be simpler. Also, if you don't need the in-kernel
>> interface, the kernel would not need to know the details of the rpmb
>> protocol at all.
>
> My major interest is the in-kernel protocol the user space API was more intended for debug, but I found it would be even more useful.
> The store type  access is very RPMB specific  for both MMC and UFS and needs to do special operations for RPMB so I don't see how this awareness can be removed or I'm not reading your proposal correctly.

The interface we use specifies reliable-write, write and read
operations on an rpmb partition. I don't think you need to know more
than this in either mmc or ufs. I have not seen the ufs spec, but
based on your code it looks like reliable-write and write can map to
the same command there.

> Accessing RPMB is a multiple stepsoperation, the steps can be driven from the user space as done in EMMC ioctl but hidning would reduce the number of system calls and amount of data passed,
> in some sense the same as in the  new mmc  MMC_IOC_MULTI_CMD is trying ot achive.
>
The main purpose of using the MMC_IOC_MULTI_CMD is not to reduce the
number of syscalls. It is to prevent other mmc operations from getting
interleaved with the rpmb packets. Some emmc chips will only respond
error packet if any other partitions are read from between the write
and read operation on the rpmb partition.

>> I have not tested your code, but it looks like we would have to modify the
>> storage proxy to interpret the data it currently passes through and remove all
>> RESULT_READ packets.
> Your proxy driver just sends ioctls so this wouldn't be much difference and the rpmb code on the trusty w need rewrite just rpmb_send () function,
> actually removing one set of buffer size. I will try to make that modification if it helps?
>

No I don't want you to modify the code that runs in the secure OS.
This would require additional code in boot loaders and proxy servers
using the existing MMC_IOC_MULTI_CMD command as they too would have to
interpret the packets to inject RESULT_READ packets.

> Thanks for interest and your review.
> Tomas
>
>



-- 
Arve Hjønnevåg

Re: [PATCH v4 7/8] mmc: block: register RPMB partition with the RPMB subsystem

[Adrian Hunter]

On 02/06/16 00:41, Tomas Winkler wrote:
> Register eMMC RPMB partition with the RPMB subsystem and provide
> implementation for the RPMB access operations abstracting
> actual multi step process.
> 
> V2: resend
> V3: commit message fix
> V4: Kconfig: use select RPMB to ensure valid configuration
>     Switch back to main area after RPMB access
> 
> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
> Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
> ---

<SNIP>

> +static void mmc_blk_rpmb_add(struct mmc_card *card)
> +{
> +	struct mmc_blk_data *md = dev_get_drvdata(&card->dev);
> +	struct mmc_blk_data *part_md = mmc_blk_rpmb_part_get(md);
> +	struct rpmb_dev *rdev;
> +
> +	if (!part_md)
> +		return;
> +
> +	mmc_blk_rpmb_set_dev_id(&mmc_rpmb_dev_ops, card);
> +
> +	/* RPMB blocks are written in half sectors hence '* 2' */
> +	mmc_rpmb_dev_ops.reliable_wr_cnt = card->ext_csd.rel_sectors * 2;

This looks like it does not support 8KB writes added in v5.1 spec.  Can that
be supported?

> +
> +	rdev = rpmb_dev_register(disk_to_dev(part_md->disk),
> +				  &mmc_rpmb_dev_ops);
> +	if (IS_ERR(rdev)) {
> +		pr_warn("%s: cannot register to rpmb %ld\n",
> +			part_md->disk->disk_name, PTR_ERR(rdev));
> +	}
> +}

<SNIP>

RE: [PATCH v4 7/8] mmc: block: register RPMB partition with the RPMB subsystem

[Winkler, Tomas]

> 2;
> 
> This looks like it does not support 8KB writes added in v5.1 spec.  Can that be
> supported?

You are right there is not support for 8K packet, but the change should be simple. 
I will update the patch.
Thanks
Tomas